Multi-physics coupled sgt mos reliability analysis platform and method

Abstract

The application provides a multi-physical field coupled SGT MOS reliability analysis platform and method, relates to the technical field of semiconductors, and the platform comprises: a structure layer splitting module which splits the structure layer of an SGT MOS device; an independent coefficient matrix construction module which imports damage state sample data under each physical field; a coupled physical field definition module which defines coupled physical fields; a coupled coefficient matrix construction module which imports joint damage state sample data under the coupled physical fields; and a reliability calculation module which constructs a reliability calculation module according to the independent influence coefficient matrix and the coupled influence coefficient matrix and outputs a first reliability calculation result. The application can solve the technical problem that the accuracy of SGT MOS reliability evaluation is low due to the neglect of the coupling effect between multiple physical fields in the prior art, and improves the accuracy of reliability evaluation through independent influence analysis and coupled influence analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor technology, and particularly relates to a SGT MOS reliability analysis platform and method based on multi-physical field coupling. BACKGROUND

[0002] SGT MOS devices have been widely used in various electronic devices due to their high switching frequency, low conduction loss and high current carrying capacity. However, in the actual working environment, SGT MOS devices usually face complex working conditions, resulting in their interaction with multiple physical fields. The existing SGT MOS device reliability analysis mainly focuses on the influence of a single physical field, ignoring the coupling effect between multiple physical fields, which leads to the inability to comprehensively reflect the performance of the device in the actual working environment when predicting the reliability of the device. Especially when multiple physical fields act together, since the physical fields are not independent but have complex interaction effects, the analysis results of a single physical field will usually deviate from the actual situation, thereby affecting the design optimization and reliability evaluation of SGT MOS devices.

[0003] In summary, in the prior art, there is a technical problem that the accuracy of SGT MOS reliability evaluation is low due to the fact that the coupling effect between multiple physical fields is ignored in the reliability analysis of SGT MOS devices. SUMMARY

[0004] The purpose of the present application is to provide a SGT MOS reliability analysis platform and method based on multi-physical field coupling, so as to solve the technical problem that the accuracy of SGT MOS reliability evaluation is low due to the fact that the coupling effect between multiple physical fields is ignored in the reliability analysis of SGT MOS devices in the prior art.

[0005] In view of the above problems, the present application provides a SGT MOS reliability analysis platform and method based on multi-physical field coupling.

[0006] In a first aspect, the present application provides a multi-physical field coupled SGT MOS reliability analysis platform, wherein the multi-physical field coupled SGT MOS reliability analysis platform comprises: a structure layer splitting module configured to split a structure layer of a SGT MOS device to output a plurality of structure layers; an independent coefficient matrix construction module configured to import damage state sample data of the plurality of structure layers under each physical field, and construct an independent influence coefficient matrix under each physical field based on the damage state sample data; a coupled physical field definition module configured to define a coupled physical field, wherein the coupled physical field is a physical field coupled by at least two physical fields; a coupled coefficient matrix construction module configured to import joint damage state sample data of the plurality of structure layers under the coupled physical field, and construct a coupled influence coefficient matrix under each coupled physical field based on the joint damage state sample data; and a reliability analysis module configured to construct a reliability calculation module according to the independent influence coefficient matrix and the coupled influence coefficient matrix, and output a first reliability calculation result of the SGT MOS device under a target coupled physical field according to the reliability calculation module.

[0007] Optionally, the plurality of structure layers at least include a metal gate layer, a polysilicon layer, a gate oxide layer, a channel region, and a substrate.

[0008] Optionally, the multi-physical field coupled SGT MOS reliability analysis platform further comprises: an index mapping unit configured to map each index in the damage state sample data to a same interval to output normalized damage score sample data; a damage score calculation unit configured to analyze the normalized damage score sample data, and calculate average damage score sample data of each structure layer under each physical field; an influence coefficient determination unit configured to evaluate an influence degree on the SGT MOS device according to the average damage score sample data of each structure layer under each physical field, and define an independent influence coefficient of each structure layer under each physical field; and a matrix arrangement unit configured to arrange the independent influence coefficients under a plurality of physical fields in a matrix to generate an independent influence coefficient matrix.

[0009] Optionally, the multi-physical field coupled SGT MOS reliability analysis platform further comprises: a joint damage index mapping unit configured to map each index in the joint damage state sample data to a same interval to output normalized joint damage score sample data; an enhancement factor determination unit configured to perform coupled calculation according to the normalized damage score sample data and the normalized joint damage score sample data, obtain a ratio of joint damage score sample data of each structure layer to average damage score sample data under a corresponding single physical field, and output a coupled enhancement factor of each structure layer under each physical field; and a coupled matrix determination unit configured to define a coupled influence coefficient vector based on the coupled enhancement factor to obtain a coupled influence coefficient matrix.

[0010] Optionally, the influence vector construction unit is used to construct a structural layer influence vector based on the independent influence coefficient matrix and the coupled influence coefficient matrix; the weight vector construction unit is used to construct a structural layer weight vector corresponding to the multiple structural layers based on the proportion of the historical failure frequency of the multiple structural layers in the total failure events; and the weighted connection unit is used to perform a weighted connection between the structural layer weight vector and the structural layer influence vector to construct a reliability calculation module.

[0011] Optionally, the reliability calculation module further includes a neighborhood influence calculation module, wherein the neighborhood influence calculation module includes a damage score judgment unit and a neighborhood diffusion calculation unit; the damage score judgment unit receives the normalized damage score sample data and the normalized joint damage score sample data, and obtains a combination of damage score sample data for each structural layer; according to the combination of damage score sample data for each structural layer, it filters out the combination of identified damage score sample data that is greater than a preset damage score threshold; the neighborhood diffusion calculation unit uses the combination of identified damage score sample data to calculate the diffusion influence coefficient of each structural layer on the diffusion of neighboring structural layers, and constructs a neighborhood diffusion influence coefficient matrix.

[0012] Optionally, the second reliability calculation unit is used for the neighborhood influence calculation module to obtain the second reliability calculation result through the neighborhood diffusion influence coefficient matrix; the calculation result update unit is used to update the first reliability calculation result according to the second reliability calculation result.

[0013] Optionally, the three-dimensional model building module is used to connect the finite element simulation system and the semiconductor multiphysics simulation tool to build a three-dimensional multilayer structure model of the SGT MOS device; the damage analysis module is used to load a single physical field and coupled physical fields onto the three-dimensional multilayer structure model, obtain simulated response data of multiple structural layers, and extract damage state sample data and joint damage state sample data from the simulated response data of multiple structural layers.

[0014] Optionally, a single physical field and coupled physical fields are loaded onto the three-dimensional multilayer structure model by setting nonlinear boundary conditions; wherein, the nonlinear boundary conditions include one or more of electric field loading boundary conditions, thermal field loading boundary conditions, stress field loading boundary conditions, and irradiation field loading boundary conditions.

[0015] In a second aspect, the present application also provides a SGT MOS reliability analysis method based on multi-physical field coupling, wherein the SGT MOS reliability analysis method based on multi-physical field coupling comprises: performing structural layer splitting on a SGT MOS device to output a plurality of structural layers; importing damage state sample data of the plurality of structural layers under each physical field, and constructing an independent influence coefficient matrix under each physical field based on the damage state sample data; defining a coupled physical field, wherein the coupled physical field is a physical field generated by coupling of at least two physical fields; importing joint damage state sample data of the plurality of structural layers under the coupled physical field, and constructing a coupled influence coefficient matrix under a plurality of coupled physical fields based on the joint damage state sample data; constructing a reliability calculation module according to the independent influence coefficient matrix and the coupled influence coefficient matrix, and outputting a first reliability calculation result of the SGT MOS device under a target coupled physical field according to the reliability calculation module.

[0016] One or more technical solutions provided in the present application have at least the following technical effects or advantages: a structural layer splitting module is used to perform structural layer splitting on a SGT MOS device to output a plurality of structural layers; an independent coefficient matrix construction module is used to import damage state sample data of the plurality of structural layers under each physical field, and construct an independent influence coefficient matrix under each physical field based on the damage state sample data; a coupled physical field definition module is used to define a coupled physical field, wherein the coupled physical field is a physical field generated by coupling of at least two physical fields; a coupled coefficient matrix construction module is used to import joint damage state sample data of the plurality of structural layers under the coupled physical field, and construct a coupled influence coefficient matrix under a plurality of coupled physical fields based on the joint damage state sample data; and a reliability analysis module is used to construct a reliability calculation module according to the independent influence coefficient matrix and the coupled influence coefficient matrix, and output a first reliability calculation result of the SGT MOS device under a target coupled physical field according to the reliability calculation module. That is, by performing structural layer splitting on a SGT MOS device, constructing an independent influence coefficient matrix under each physical field, defining a coupled physical field, constructing a coupled influence coefficient matrix under a plurality of coupled physical fields, and combining the independent influence coefficient matrix and the coupled influence coefficient matrix to construct a reliability calculation module, the reliability of the SGT MOS device under a target coupled physical field is calculated, and the accuracy and efficiency of SGT MOS device reliability evaluation are improved.

[0017] The above description is only a summary of the technical solutions of the present application. In order to enable the technical means of the present application to be more clearly understood, and to be implemented according to the contents of the description, and in order to enable the above and other purposes, characteristics and advantages of the present application to be more apparent and easy to understand, the following specific embodiments of the present application are described. It should be understood that the contents described in this part are not intended to identify the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and other drawings can also be obtained by the provided drawings without creative labor for those skilled in the art.

[0019] Figure 1 The structure schematic diagram of the SGT MOS reliability analysis platform of the multi-physical field coupling of the present application.

[0020] Figure 2 The flowchart of the SGT MOS reliability analysis method of the multi-physical field coupling of the present application.

[0021] Explanation of reference signs: structure layer splitting module 11, independent coefficient matrix construction module 12, coupled physical field definition module 13, coupled coefficient matrix construction module 14, reliability analysis module 15. DETAILED DESCRIPTION

[0022] The present application provides a multi-physical field coupling SGT MOS reliability analysis platform and method, which solves the technical problem that the accuracy of SGT MOS reliability evaluation is low due to the neglect of the coupling effect between multiple physical fields in the reliability analysis of SGT MOS devices in the prior art. The SGT MOS device is split into structure layers; the independent influence coefficient matrix under multiple physical fields is constructed; the coupled physical field is defined; the coupled influence coefficient matrix under multiple coupled physical fields is constructed; the reliability calculation module is constructed by combining the independent influence coefficient matrix and the coupled influence coefficient matrix, the reliability of the SGT MOS device under the target coupled physical field is calculated, and the SGT MOS device reliability evaluation precision and efficiency are improved.

[0023] Hereinafter, the technical solutions in the present application will be described clearly and completely with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited to the described example embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application. In addition, it should be noted that, for the convenience of description, only parts related to the present application are shown in the drawings, not all.

[0024] Embodiment one, please refer to the attached Figure 1 The present application provides a multi-physical field coupled SGT MOS reliability analysis platform, wherein the multi-physical field coupled SGT MOS reliability analysis platform is used to implement the steps of the multi-physical field coupled SGT MOS reliability analysis method, and the multi-physical field coupled SGT MOS reliability analysis platform comprises:

[0025] The structural layer splitting module 11 is configured to split the structural layers of the SGT MOS device and output a plurality of structural layers.

[0026] Further, the structural layer splitting module 11 in the multi-physical field coupled SGT MOS reliability analysis platform is further configured to: the plurality of structural layers at least include a metal gate layer, a polysilicon layer, a gate oxide layer, a channel region and a substrate.

[0027] Specifically, SGT MOS is a split gate MOSFET, which is a metal oxide semiconductor field effect transistor designed with split gate. There are two gate regions (such as control gate and selection gate) on the gate structure, which control different parts respectively. Compared with traditional MOSFET, it has higher precise control ability, can reduce parasitic effects, and performs better especially in high frequency or high power applications. Through the split gate design, SGT MOS can improve performance, reduce energy loss, and improve work efficiency.

[0028] In the structural layer splitting of SGT MOS devices, first determine the different levels contained in the SGT MOS device, at least including the metal gate layer, the polysilicon layer, the gate oxide layer, the channel region and the substrate. Each layer has different material characteristics and electrical properties. Structural layer splitting refers to the decomposition of each level of the SGT MOS device, and each physical field will have different effects on different layers of the SGT MOS device. Through structural layer splitting, more detailed reliability evaluation can be performed. The characteristics of each layer need to be analyzed in detail by simulation tools (such as finite element analysis), and the SGT MOS device is split into metal gate layer, polysilicon layer, gate oxide layer, channel region and substrate, etc., and the physical properties of each layer are calibrated. The electrical properties of the metal gate layer, such as conductivity and electric field distribution; the conductivity of the polysilicon layer and the gate control ability; the influence of the gate oxide layer on the performance of the electric field isolation and thickness; the carrier mobility and conductivity change of the channel region; the thermal conductivity and material stress distribution of the substrate.

[0029] The metal gate layer is the gate region of the SGT MOS device, usually composed of metal materials (such as aluminum or molybdenum), responsible for controlling the flow of electrons, forming an electric field between the semiconductor channel region, and then controlling the opening and closing of the conduction channel. The polysilicon layer is usually used in the gate part of the MOSFET as a conductive material, located between the metal gate layer and the gate oxide layer. Compared with the metal gate layer, the polysilicon layer has better control ability and lower resistance, and can provide the ability to accurately control current flow. The gate oxide layer is an oxide layer between the gate and the channel, usually silicon dioxide, and its thickness directly affects the switching performance of the device. The gate oxide layer is responsible for isolating the gate and the channel to prevent direct current flow, while controlling the influence of the electric field in the channel region. The electric field between the metal gate layer and the gate oxide layer enables the SGT MOS to accurately control the switching state of the device. The channel region is the region connecting the source and the drain, where the current flows. The gate voltage controls the conductivity of this region, thereby adjusting the current between the source and the drain. The substrate is the base of the entire device, usually silicon material, providing mechanical support and electrical connection for the SGT MOS device.

[0030] By structurally splitting the SGT MOS device and separately and coupling analyzing the multi-physical field influence of each layer, the reliability of the SGT MOS device in the actual working environment can be accurately evaluated. For example, after structural layer splitting, the metal gate layer (thickness of 100 nm), the polysilicon layer (thickness of 200 nm), the gate oxide layer (thickness of 3 nm), the channel region (length of 1 μm, width of 0.5 μm) and the substrate (thickness of 500 μm) are obtained.

[0031] An independent coefficient matrix construction module 12 is configured to import damage state sample data of the plurality of structure layers under each physical field, and construct an independent influence coefficient matrix under a plurality of physical fields based on the damage state sample data.

[0032] Further, the multi-physical field coupled SGT MOS reliability analysis platform further comprises a three-dimensional model construction module configured to connect a finite element simulation system and a semiconductor multi-physical field simulation tool to construct a three-dimensional multi-layer structure model of the SGT MOS device; and a damage analysis module configured to load a single physical field and a coupled physical field to the three-dimensional multi-layer structure model, obtain a plurality of structure layer simulation response data, and extract damage state sample data and joint damage state sample data from the plurality of structure layer simulation response data.

[0033] Further, the multi-physical field coupled SGT MOS reliability analysis platform further comprises: the three-dimensional multi-layer structure model is loaded with a single physical field and a coupled physical field by setting a nonlinear boundary condition; wherein the nonlinear boundary condition comprises one or more of an electric field loading boundary condition, a thermal field loading boundary condition, a stress field loading boundary condition, and an irradiation field loading boundary condition.

[0034] Specifically, the finite element simulation system is a computer program for numerical analysis in the fields of engineering and physics, which divides complex structure or field problems into smaller, more easily analyzed elements, and calculates their behavior under different physical conditions through numerical methods. The semiconductor multi-physical field simulation tool is a simulation tool specifically designed to simulate the effects of multiple physical fields (such as electric field, thermal field, force field, irradiation field, etc.) on semiconductor devices, which can consider the interaction of multiple physical effects in the same model, thereby providing more comprehensive and accurate performance evaluation.

[0035] Connecting the finite element simulation system and the semiconductor multi-physical field simulation tool, a three-dimensional multi-layer structure model of the SGT MOS device is constructed. This means that the finite element simulation system and the semiconductor multi-physical field simulation tool are integrated into a complete simulation process, ensuring joint simulation of different physical effects of the SGT MOS device, thereby providing more accurate evaluation and prediction. The finite element simulation system is used to handle structural mechanics and heat transfer problems, while the semiconductor multi-physical field simulation tool can simulate the interaction of multiple physical fields such as electric field, thermal field, force field, and irradiation field.

[0036] A three-dimensional multi-layer model is established using a semiconductor multi-physics simulation tool, based on the design parameters of the SGT MOS device. The model includes multiple structural layers such as the metal gate layer, polysilicon layer, gate oxide layer, channel region, and substrate. During the construction of the three-dimensional structure, geometric refinement is required, which involves dividing each layer of the device into small units to improve computational accuracy. For example, the modeling of the metal gate layer and gate oxide layer will be relatively fine, as they directly affect the distribution of the electric field; while the modeling of the substrate and channel region needs to consider thermal management and stress distribution more carefully. The three-dimensional multi-layer structure model not only reflects the physical structure of the device, but also enables multi-physics analysis to evaluate the interaction between layers and their impact on device performance.

[0037] The three-dimensional multi-layer structure model is loaded with single physical fields and coupled physical fields by setting nonlinear boundary conditions. Nonlinear boundary conditions refer to conditions on the boundary that are not linear, i.e., they cannot be represented as a simple linear combination of physical quantities, including one or more of the following: electric field loading boundary conditions, thermal field loading boundary conditions, stress field loading boundary conditions, and irradiation field loading boundary conditions. For example, electric fields, thermal fields, or stress fields may exhibit nonlinear characteristics at different intensities. In the multi-physics simulation of semiconductor devices, nonlinear boundary conditions are commonly used to simulate the complex behavior of physical fields as conditions change, such as the impact of temperature changes on the conductivity of materials, or the impact of electric fields on the internal structure of SGT MOS devices as voltage changes.

[0038] Electric field loading boundary conditions typically specify the application of an electric field or voltage at specific locations in the SGT MOS device, such as the metal gate layer, electrodes, or substrate; thermal field loading boundary conditions involve simulating the thermal effects of the SGT MOS device during operation due to current flow, power loss, etc. by applying heat sources or controlling temperature; stress field loading boundary conditions refer to stress effects caused by external or internal factors, particularly under conditions of temperature change or high power; and irradiation field loading boundary conditions are typically used to simulate the impact of high-energy radiation (such as ultraviolet light, electromagnetic radiation, etc.) on the SGT MOS device.

[0039] A single physical field refers to the impact of a single physical phenomenon on the performance of an SGT MOS device, such as only loading an electric field, a thermal field, or a stress field. Under the loading of a single physical field, the simulation tool will only consider the effect of that physical phenomenon (e.g., the effect of voltage changes on the electric field) on the SGT MOS device. Coupled physical fields refer to the case where multiple physical phenomena (such as electric fields, thermal fields, and force fields) interact with each other. During the operation of the device, the physical fields are usually not independent, but have interactive effects. For example, when loading a single physical field, the electric field distribution is determined by applying a given gate voltage and electrode voltage to load the electric field. When a 5V voltage is applied to the gate, the electric field strength distribution between the metal gate layer and the gate oxide layer is calculated, and its effect on the channel region is analyzed. When loading a thermal field, the thermal effect caused by the current passing through is simulated, and the current density is set to 10 A / cm². The temperature distribution inside the device is analyzed, assuming that the temperature near the gate oxide layer rises to 100°C, and the effect of this temperature on the performance of the device is calculated. When loading coupled physical fields, heat is generated due to the flow of current, and an electro-thermal coupling analysis is performed. The current is set to 5A, the thermal effect is calculated through the current, and the temperature is adjusted, further affecting the distribution of the electric field. The temperature change caused by the thermal field will cause the material to expand or contract, thereby generating stress in the structure and affecting the stability of the device. The mechanical response of the substrate or other materials to temperature changes is simulated through thermal-mechanical coupling simulation. Assuming that the thermal expansion coefficient of the substrate is 11 × 10 -6 K -1 , the stress change of the substrate is calculated when the temperature rises to 120°C.

[0040] After loading single physical fields and coupled physical fields, multiple structural layer simulation response data is obtained, including the response data of each structural layer under the action of these physical fields, such as current density, electric field strength, temperature distribution, stress distribution, etc. Through the analysis of the separate loading of each physical field, damage state sample data is extracted. By considering the comprehensive effect of multiple physical fields under the action of coupling, joint damage state sample data is extracted, reflecting the overall damage state of the SGT MOS device under different environmental conditions under the interaction of multiple physical fields.

[0041] Simulation response data refers to the results calculated and output by a three-dimensional multi-layer structure model after loading different physical fields (such as electric fields, thermal fields, etc.). Damage state sample data is usually obtained by calculating the effect of electric fields, thermal fields, stress fields, or irradiation fields on the device structure, and converting these effects into specific damage assessment data, such as current leakage, stress change, etc. For example, the leakage current of the gate oxide layer may increase with the increase of temperature, thereby causing damage to the state. The leakage current of the gate oxide layer increases from 5 × 10 -17 A to 6 × 10 -17A, indicating that damage occurs under the action of thermal field. The joint damage state sample data refers to the comprehensive damage data of each layer of the device under the coupling action of multiple physical fields. Through the joint simulation of multiple physical fields, more accurate damage prediction results are obtained, considering the comprehensive influence of the interaction of multiple physical fields (such as the coupling of electric field and thermal field) on the device. For example, under the joint action of electric field and thermal field, the current density of the metal gate layer increases by 25%, and the leakage current of the gate oxide layer increases by 20%. By setting nonlinear boundary conditions, loading single physical field and coupled physical field, and extracting damage state sample data and joint damage state sample data, accurate and reliable reliability evaluation of SGT MOS device can be realized.

[0042] Further, the independent coefficient matrix construction module 12 in the multi-physical field coupled SGT MOS reliability analysis platform is further used for:

[0043] The index mapping unit is configured to map each index in the damage state sample data to the same interval, and output normalized damage score sample data. The damage score calculation unit is configured to analyze the normalized damage score sample data, and calculate average damage score sample data of each structure layer under each physical field. The influence coefficient determination unit is configured to evaluate the influence degree of each structure layer under each physical field according to the average damage score sample data of each structure layer under each physical field, and define an independent influence coefficient of each structure layer under each physical field. The matrix arrangement unit is configured to arrange the independent influence coefficients under multiple physical fields in a matrix to generate an independent influence coefficient matrix.

[0044] Specifically, the damage state sample data is the damage condition data of each structure layer of the SGT MOS device under the action of different physical fields, including the damage information of each layer (such as metal gate layer, gate oxide layer, channel region, etc.) of the device under the action of electric field, thermal field, force field, irradiation field, etc., such as the change of current density, temperature rise or stress increase, etc. Each index in the damage state sample data is mapped to the same interval, i.e. mapped to the same interval (such as [0, 1]), eliminating the influence of different dimensions or different ranges, so that the damage score can be compared between physical fields or structure layers. Generally, through normalization processing, the damage evaluation between different physical fields or different layers has comparability.

[0045] The average damage score of each structure layer (such as metal gate layer, polysilicon layer, gate oxide layer, channel region and substrate, etc.) under each physical field (such as electric field, thermal field, force field, irradiation field, etc.) is calculated, and the average value obtained by averaging all damage state sample data is usually used to represent the overall damage degree of the layer under the physical field. The average damage score refers to the statistical analysis of the damage score of each structure layer under each physical field, and the average damage level of the layer is calculated, which is usually the average value of multiple damage state sample data, used to represent the damage degree of the layer under the specific physical field condition. For example, for the gate oxide layer under the action of electric field, assuming that the normalized damage sample data is [0.6, 0.7, 0.8, 0.9, 1.0], the average damage score is: (0.6+0.7+0.8+0.9+1.0) / 5=0.8; for the metal gate layer under the action of thermal field, assuming that the normalized damage sample data is [0.5, 0.6, 0.7, 0.8, 0.9], the average damage score is: (0.5+0.7+0.8+0.9+1.0) / 5=0.7.

[0046] According to the average damage score sample data of each structure layer under each physical field, the influence degree on the SGT MOS device is evaluated, and the independent influence coefficient of each structure layer under each physical field is defined. The influence degree refers to the action strength of each physical field on each structure layer of the device. The influence degree is usually measured by the change of damage score. The higher the damage score, the greater the influence of the physical field on the structure layer.

[0047] Through the analysis of the average damage score sample data, the influence degree of each physical field on each structure layer can be evaluated. The independent influence coefficient refers to the influence strength of each physical field on each structure layer, and through the analysis of the average damage score sample data, the correlation between the damage score of each layer under the action of each physical field and the physical field is calculated. For example, if the average damage score of the gate oxide layer under the action of thermal field is 0.8, and the average damage score of the metal gate layer is 0.3, then the influence degree of thermal field on the gate oxide layer is greater than that on the metal gate layer, indicating that the influence of thermal field on the gate oxide layer is strong, while the influence on the metal gate layer is weak.

[0048] By evaluating the influence degree under each physical field, the independent influence coefficient of each physical field to each structure layer is defined, which represents the influence intensity of each physical field on the structure layer. The independent influence coefficients of multiple structure layers under multiple physical fields are arranged in a matrix, i.e. the independent influence coefficients are arranged according to the physical field and the structure layer, forming an independent influence coefficient matrix, the rows represent different physical fields (electric field, thermal field, force field, irradiation field, etc.), and the columns represent different structure layers (such as metal gate layer, polysilicon layer, gate oxide layer, channel region, super junction, substrate, etc.). Exemplarily, for the influence evaluation of a certain SGT MOS device, if the influence degree of the electric field on the gate oxide layer is 0.8, i.e. the influence of the electric field on the gate oxide layer is the largest, the independent influence coefficient of the electric field on the gate oxide layer is set to 1.0; if the influence degree of the thermal field on the gate oxide layer is 0.7, the independent influence coefficient of the thermal field on the gate oxide layer is set to 0.8, and the independent influence coefficient of each structure layer under each physical field is repeated to obtain the independent influence coefficient matrix (i.e. the influence weight of each physical field on each structure layer) as shown in Table 1.

[0049] Table 1 Independent influence coefficient matrix

[0050]

[0051] The independent influence coefficient matrix of Table 1 shows the influence degree of different physical fields on each layer (L1 to L6) of the SGT MOS device. Under the action of the electric field, the damage degree of the gate oxide layer is the largest, which is 1.0, i.e. the influence of the electric field on the gate oxide layer is much stronger than that on other structure layers. The influence coefficient of the thermal field on the polysilicon layer is 0.6, and the influence coefficient on the gate oxide layer is 0.8, indicating that the influence of the thermal field on the gate oxide layer is strong, but the influence on other structure layers is weak. The influence coefficient of the force field on the polysilicon layer is 0.7, indicating that the force field has a great influence on the polysilicon layer, which may be due to the mechanical stress caused by thermal expansion. The influence coefficient of irradiation on the substrate layer is 0.5, indicating that the influence of surface irradiation on the substrate layer is large, especially in the radiation environment, the damage of the substrate layer is more significant than that of other layers. By quantifying the influence of each physical field on each structure layer, the damage of the device under complex working conditions is accurately evaluated, thereby improving the accuracy of reliability evaluation. By defining the independent influence coefficient, the damage state of each structure layer under different physical fields is comprehensively understood, which helps to identify the physical field and structure layer with the greatest influence, optimize the design parameters, and improve the device performance.

[0052] The coupling physical field defining module 13 is configured to define a coupling physical field, wherein the coupling physical field is a physical field coupled by at least two physical fields.

[0053] In particular, coupled physical fields refer to the physical phenomenon that at least two physical fields (such as electric field, thermal field, force field, irradiation, etc.) interact with each other and jointly act. In the multi-physical field simulation of SGT MOS devices, the interaction between each physical field is very important, because considering each physical field alone may not truly reflect the performance of the device in the actual working environment. The key feature of coupled physical fields is that the physical fields are not independent, but interact with each other and jointly determine the behavior of the system. For example, the interaction between electric field and thermal field, the current flow causes heat generation, and the temperature rise affects the electric field distribution, which is a typical coupled physical field.

[0054] Define the physical fields that affect the performance of SGT MOS devices, for example, the electric field is generated by the gate voltage and the source-drain voltage; the thermal field is generated by the power loss caused by current flow; the force field may be caused by stress due to temperature change or external load. Determine the interaction between each physical field. For example, current flow causes heat generation (electric-thermal coupling), and temperature change affects the path of current flow (thermal-electric coupling); stress (force field) may affect the electric field distribution, especially in high power or high voltage cases (electric-force coupling).

[0055] Load physical fields such as electric field, thermal field, force field, etc. using finite element analysis tools, and perform simulation calculations under each physical field. Link each physical field through the coupling relationship. In electric-thermal coupling simulation, the simulation tool needs to consider current flow and heat generation at the same time, and adjust the electric field distribution according to the temperature distribution; in electric-force coupling simulation, the thermal expansion of the device may change the distribution of the electric field, which needs to be simulated through multi-physical field coupling. For each physical field, appropriate boundary conditions need to be set. For example, when loading the electric field, the gate voltage can be set, when loading the thermal field, the power loss and temperature change can be considered; when loading the stress field, the boundary conditions of the material thermal expansion coefficient and external force need to be set.

[0056] By defining and simulating coupled physical fields, the interaction between multiple physical fields and their impact on SGT MOS devices can be accurately evaluated. By simulating and analyzing the impact of coupled physical fields, the reliability of the device can be accurately evaluated, and the design and manufacturing process of the device can be optimized.

[0057] The coupling coefficient matrix construction module 14 is configured to import joint damage state sample data of the plurality of structure layers under the coupled physical fields, and construct a coupling influence coefficient matrix under the plurality of coupled physical fields based on the joint damage state sample data.

[0058] Further, the coupling coefficient matrix construction module 14 in the multi-physical field coupled SGT MOS reliability analysis platform is also used for: a joint damage index mapping unit, configured to map each index in the joint damage state sample data to the same interval, and output normalized joint damage score sample data; an enhancement factor determination unit, configured to perform coupling calculation according to the normalized damage score sample data and the normalized joint damage score sample data, obtain a ratio of joint damage score sample data of each structure layer to average damage score sample data under the action of a corresponding single physical field, and output a coupling enhancement factor of each structure layer under each physical field; and a coupling matrix determination unit, configured to define a coupling influence coefficient vector based on the coupling enhancement factor, and obtain a coupling influence coefficient matrix.

[0059] Specifically, the joint damage state sample data refers to damage condition data of each structure layer of a device under the joint action of multiple physical fields (such as electric field, thermal field, force field, etc.), which reflects the comprehensive influence of the coupling action between the physical fields on the structure layer, and can provide more accurate damage evaluation than a single physical field. Mapping each index in the joint damage state sample data to the same interval, that is, mapping each index in the joint damage state sample data to the same standard range (such as [0, 1]), makes the damage data under the action of different physical fields comparable and eliminates the dimensional difference of data of different physical fields or different structure layers. The normalized joint damage score sample data is obtained by standardizing the joint damage state sample data, so that all damage scores are in the same dimension and range (usually [0, 1]), ensuring that the comparison between different physical fields and different structure layers is consistent and fair. Coupling calculation refers to ratio calculation of the normalized damage score sample data and the normalized joint damage score sample data, to analyze the relationship between the comprehensive influence of the coupling action of multiple physical fields on each structure layer of the device and the influence of the action of a single physical field. For each structure layer, the ratio between the joint damage score under the action of multiple physical fields and the average damage score under the action of a single physical field is calculated, reflecting the enhancement effect of the coupling action of multiple physical fields relative to the action of a single physical field.

[0060] According to the coupling enhancement factor of each physical field to each structure layer, a coupling influence coefficient vector is defined, which includes the coupling influence coefficient of each physical field to each structure layer, and is a coefficient set quantifying the coupling effect of each physical field. The coupling influence coefficient vector is a vector containing the coupling influence coefficient of each physical field to each structure layer. Each element represents the coupling influence intensity of a certain structure layer of the device under the action of a specific physical field. The coupling influence coefficient vector is used to identify the influence degree of different physical fields on different layers of the device under the action of multiple physical field coupling. The coupling influence coefficient matrix is obtained by arranging the coupling influence coefficient vector of each physical field to each structure layer in rows. The rows of the matrix represent different physical fields (such as electric field, thermal field, force field, irradiation field, etc.), and the columns represent different structure layers (such as metal gate layer, polysilicon layer, gate oxide layer, channel region, super junction, substrate, etc.). For example, assuming that the joint damage score of the super junction is 0.75, the average damage score under the action of the thermal field is 0.7, and the average damage score under the action of the force field is 0.6. The coupling enhancement factor is calculated as follows: coupling enhancement factor (force field) = 1.25, which means that the damage degree of the super junction is enhanced by 1.25 times when the force field and other physical fields act together; the coupling enhancement factor (thermal field) is 1.07, which means that the damage degree of the super junction is enhanced by 1.07 times when the thermal field and other physical fields act together. For other coupled physical fields, repeat the foregoing calculation process, and the coupling influence coefficient matrix is shown in Table 2:

[0061] Table 2 Coupling influence coefficient matrix

[0062]

[0063] Arranging the coupling influence coefficient vector of each physical field to each structure layer in rows generates a coupling influence coefficient matrix, which provides the influence intensity of multiple physical fields on each structure layer. Each matrix element represents the coupling influence coefficient of a specific physical field on a specific structure layer. By calculating the coupling enhancement factor and generating the coupling influence coefficient matrix, the comprehensive influence of multiple physical fields on the structure layers of the device is quantified, further improving the accuracy of the multi-physical field coupling analysis. Through coupling calculation and the coupling influence coefficient matrix, the reliability evaluation accuracy of the device in a multi-physical field environment is improved.

[0064] The reliability analysis module 15 is configured to construct a reliability calculation module according to the independent influence coefficient matrix and the coupling influence coefficient matrix, and output a first reliability calculation result of the SGT MOS device under a target coupled physical field according to the reliability calculation module.

[0065] Further, the reliability analysis module 15 in the SGT MOS reliability analysis platform of the multi-physical field coupling is further configured to:

[0066] An influence vector construction unit is configured to construct a structure layer influence vector based on the independent influence coefficient matrix and the coupled influence coefficient matrix; a weight vector construction unit is configured to construct a structure layer weight vector corresponding to each structure layer based on the proportion of the plurality of structure layer historical failure frequencies in total failure events; and a weighted connection unit is configured to perform weighted connection on the structure layer weight vector to the structure layer influence vector to construct a reliability calculation module.

[0067] Specifically, a structure layer influence vector is constructed based on the independent influence coefficient matrix and the coupled influence coefficient matrix, the two matrices are combined, and the influence of each physical field is weighted and summed to obtain a comprehensive influence coefficient of each structure layer, i.e., the structure layer influence vector. The structure layer influence vector is a vector calculated based on the independent influence coefficient matrix and the coupled influence coefficient matrix, representing the overall influence degree of each structure layer after considering the influence of all physical fields, and reflecting the comprehensive action of multiple physical fields on each layer of the device.

[0068] The structure layer historical failure frequency refers to the number of times that each structure layer fails in past failure events, which is recorded through historical data and reflects the failure probability of different structure layers in actual use. The total failure events of multiple structure layers are collected and analyzed, and the historical failure frequency data of each structure layer is also collected. The proportion of each structure layer in the total failure events is calculated through the ratio of the failure frequency of each structure layer to the total failure frequency of all structure layers. The proportion of the structure layer failure frequency in the total failure events refers to the ratio of the failure frequency of each structure layer to the total failure frequency of all structure layers, reflecting the relative importance and contribution of a certain structure layer in all failure events. For example, assuming that in the past 50 failure events: the metal gate layer fails 10 times, the polysilicon layer fails 5 times, the gate oxide layer fails 15 times, the channel region fails 8 times, the super junction fails 7 times, and the substrate layer fails 5 times. For example, the proportion of the metal gate layer is 0.2, the proportion of the polysilicon layer is 0.1, the proportion of the gate oxide layer is 0.3, the proportion of the channel region is 0.16, the proportion of the super junction is 0.14, and the proportion of the substrate layer is 0.1.

[0069] The structure layer weight vector is a vector constructed based on the proportion of the historical failure frequency of each structure layer, and each element of the vector represents the weight of the structure layer in all failure events. The higher the weight, the greater the influence of the structure layer on the overall device reliability, and vice versa. The structure layer weight vector is constructed based on the proportion of the historical failure frequency of each structure layer. For example, based on the failure proportions calculated in the foregoing, the weight vector of the structure layer is [0.20, 0.10, 0.30, 0.16, 0.14, 0.10].

[0070] According to the structure layer weight vector, the structure layer influence vector is weighted and connected, that is, each element in the influence vector is multiplied by the corresponding weight, to construct a reliability calculation module. The reliability calculation module is obtained by combining the structure layer influence vector and the structure layer weight vector, and performing weighted calculation, and is used for finally evaluating the overall reliability of the SGT MOS device. The influence degree and importance of each structure layer are combined to generate a comprehensive reliability prediction result. By constructing the reliability calculation module, the reliability of the SGT MOS device is quantitatively evaluated, the most critical structure layer in the device can be identified, and thus the SGT MOS device can be optimized and designed to improve the long-term stability and reliability of the SGT MOS device.

[0071] The target coupled physical field refers to a specific combination of physical fields of interest, which have interaction and coupling between them. By performing reliability analysis on the SGT MOS device in the target coupled physical field, a first reliability calculation result of the SGT MOS device in the target coupled physical field is obtained. The first reliability calculation result is a device overall reliability evaluation obtained based on multi-physical field coupling calculation, which is usually expressed as a failure probability or service life of the device. By coupling the influence of different physical fields, the comprehensive influence under the interaction of multiple physical fields is considered, thereby improving the accuracy of the device reliability evaluation.

[0072] Further, the reliability analysis module 15 in the multi-physical field coupled SGT MOS reliability analysis platform is also used for:

[0073] The reliability calculation module further includes a neighborhood influence calculation module, wherein the neighborhood influence calculation module includes a damage score judgment unit and a neighborhood diffusion calculation unit; the damage score judgment unit receives the normalized damage score sample data and the normalized joint damage score sample data, obtains a damage score sample data combination of each structure layer; according to the damage score sample data combination of each structure layer, an identified damage score sample data combination greater than a preset damage score threshold is screened; and the neighborhood diffusion calculation unit calculates a diffusion influence coefficient of each structure layer caused by neighborhood structure layer diffusion by using the identified damage score sample data combination, to construct a neighborhood diffusion influence coefficient matrix.

[0074] Specifically, when the damage score of a certain structure layer exceeds the preset threshold, the damage of this structure layer will affect its adjacent structure layer, which is caused by the damage propagation between structure layers, and usually needs to be quantified by calculating the diffusion influence coefficient. According to the preset damage score threshold, it is judged whether the damage score of each structure layer exceeds the threshold. If the damage score of a certain structure layer is greater than the threshold, the layer is marked as a damage state, and the diffusion influence on the adjacent structure layer needs to be further calculated. For example, if the preset threshold is 0.8 and the joint damage score of the gate oxide layer under the coupling effect of the electric field and the thermal field is calculated to be 0.85, the damage score of the gate oxide layer exceeds the threshold, and is marked as a damage state. Once it is determined that the damage score of a certain structure layer exceeds the threshold, the neighborhood diffusion calculation unit will calculate the diffusion influence coefficient of the structure layer on the adjacent structure layer. The diffusion influence coefficient represents how the damage of a structure layer affects its adjacent structure layer, reflecting the propagation effect of damage between structure layers.

[0075] The neighborhood influence calculation module is part of the reliability calculation module, which is specially used to calculate the propagation effect of damage between structure layers. The neighborhood influence calculation module includes two main components: a damage score judgment unit and a neighborhood diffusion calculation unit, which work together to quantify the diffusion influence between adjacent structure layers and provide key data for reliability evaluation. The damage score judgment unit is a submodule of the neighborhood influence calculation module, which is used to receive normalized damage score sample data and normalized joint damage score sample data, and judge whether there is obvious damage in each structure layer according to the preset damage score threshold. The damage score judgment unit will screen out the structure layers whose damage scores exceed the threshold, and mark these structure layers as damage states.

[0076] The neighborhood diffusion calculation unit is another submodule of the neighborhood influence calculation module, which calculates how the damage of the damage structure layer screened out by the damage score judgment unit affects the damage state of the adjacent structure layer, quantifies the propagation effect of damage between structure layers through the diffusion model, calculates the diffusion influence coefficient of each structure layer on the adjacent layer, and finally constructs the neighborhood diffusion influence coefficient matrix.

[0077] The damage score judgment unit receives normalized damage score sample data and normalized joint damage score sample data, which contains the damage state of each structure layer of the device under different working conditions. According to the preset damage score threshold, it is judged whether the damage score of each structure layer exceeds the threshold. If the damage score of a certain structure layer exceeds the threshold, it is marked as a damage state, otherwise it is marked as a non-damage state. Once the damage score judgment unit screens out the structure layer in the damage state, the neighborhood diffusion calculation unit will calculate the diffusion influence coefficient of the damage state structure layer on its adjacent structure layer, quantifying the propagation effect of damage between different structure layers.

[0078] By calculating the diffusion influence coefficient between all structure layers, the neighborhood diffusion influence coefficient matrix is obtained. The rows and columns of the matrix represent different structure layers, and each element represents the diffusion influence coefficient of a certain structure layer on its adjacent structure layer. For example, assuming that the gate oxide layer damage score (0.85) exceeds the threshold value (0.85), the damage score judgment unit marks it as a damage state. The neighborhood diffusion calculation unit calculates the diffusion influence coefficient of the gate oxide layer on its adjacent structure layers (such as the channel region and the substrate layer). Assuming that the diffusion influence coefficient of the gate oxide layer on the channel region is 0.6, and the influence coefficient on the substrate layer is 0.4, which represents the influence of the damage of the gate oxide layer on other structure layers. The neighborhood diffusion influence coefficient matrix is composed of multiple diffusion influence coefficients, reflecting the damage propagation effect of each structure layer on its adjacent structure layers. The rows and columns of the matrix represent different structure layers, and the numerical value of each cell represents the diffusion influence coefficient of a certain structure layer on its adjacent layer. After the neighborhood diffusion influence coefficient matrix is constructed, the reliability calculation module will combine these data to calculate the overall reliability of the device under the target coupled physical field, considering the damage state of each structure layer, the neighborhood diffusion effect and other influencing factors, and finally output the reliability evaluation result of the device.

[0079] By calculating the diffusion influence coefficient, the propagation effect of damage between structure layers is accurately simulated, the interaction effect between neighborhood layers is considered, the influence of multi-physical field and interlayer interaction is fully reflected, and the accuracy of reliability evaluation is further improved. By identifying which structure layer has a greater diffusion influence on adjacent layers, measures can be taken to reduce damage diffusion, thereby optimizing the design of the SGT MOS device and improving the stability and long-term reliability of the SGT MOS device.

[0080] Further, the reliability analysis module 15 in the multi-physical field coupled SGT MOS reliability analysis platform is also used for:

[0081] The second reliability calculation unit is used for the neighborhood influence calculation module to obtain a second reliability calculation result through the neighborhood diffusion influence coefficient matrix; and the calculation result updating unit is used for updating the first reliability calculation result according to the second reliability calculation result.

[0082] Specifically, the neighborhood influence calculation module is used to calculate the diffusion influence between each structure layer and finally output the calculation result of the influence, and the neighborhood diffusion influence coefficient matrix is used to quantify the damage diffusion effect of one structure layer on the adjacent structure layer. The neighborhood influence calculation module uses the calculated neighborhood diffusion influence coefficient matrix to reflect the diffusion influence of damage from a certain structure layer to the adjacent layer. Each element represents the diffusion influence coefficient of a certain structure layer on its neighborhood layer. The neighborhood influence calculation module uses the neighborhood diffusion influence coefficient matrix to obtain the second reliability calculation result, which is calculated based on the influence degree of each structure layer on the adjacent structure layer, and comprehensively reflects the reliability level of the device under a specific working condition. The second reliability calculation result is the result calculated by the neighborhood influence calculation module, which considers the diffusion influence between structure layers and reflects the reliability of the device under the condition that there is damage propagation between structure layers.

[0083] After obtaining the second reliability calculation result, the first reliability calculation result will be updated according to the second reliability calculation result, that is, the neighborhood diffusion effect is introduced into the reliability calculation, so that the evaluation result is more accurate. For example, assuming that the first reliability calculation result is the reliability score of the gate oxide layer 0.85, and the diffusion influence coefficient of the gate oxide layer on the channel region is 0.6, then the influence of the gate oxide layer is 0.85*0.6=0.51, and the influence of the gate oxide layer on the substrate layer is 0.85*0.4=0.34.

[0084] The second reliability calculation result is applied to the first reliability calculation result for updating to obtain the final reliability evaluation result, which accurately reflects the influence of multi-physical field coupling and interlayer damage diffusion effect on the device. By introducing the neighborhood diffusion influence coefficient matrix, the damage propagation effect between structure layers is considered, and the accuracy of reliability evaluation is effectively improved, especially in the damage propagation process under the action of multiple physical fields.

[0085] In summary, the SGT MOS reliability analysis platform provided by the present application has the following technical effects:

[0086] The structural layer splitting module is used for splitting the structural layers of the SGT MOS device to output a plurality of structural layers; the independent coefficient matrix construction module is used for importing damage state sample data of the plurality of structural layers under each physical field, and constructing independent influence coefficient matrices under a plurality of physical fields based on the damage state sample data; the coupled physical field definition module is used for defining a coupled physical field, wherein the coupled physical field is a physical field coupled by at least two physical fields; the coupled coefficient matrix construction module is used for importing joint damage state sample data of the plurality of structural layers under the coupled physical field, and constructing coupled influence coefficient matrices under a plurality of coupled physical fields based on the joint damage state sample data; and the reliability calculation module is used for constructing a reliability calculation module according to the independent influence coefficient matrices and the coupled influence coefficient matrices, and outputting a first reliability calculation result of the SGT MOS device under a target coupled physical field according to the reliability calculation module. That is, the structural layers of the SGT MOS device are split; the independent influence coefficient matrices under a plurality of physical fields are constructed; the coupled physical field is defined; the coupled influence coefficient matrices under a plurality of coupled physical fields are constructed; the independent influence coefficient matrices and the coupled influence coefficient matrices are combined to construct the reliability calculation module, and the reliability of the SGT MOS device under the target coupled physical field is calculated, thereby improving the reliability evaluation precision and efficiency of the SGT MOS device.

[0087] In the second embodiment, based on the same inventive concept as the SGT MOS reliability analysis platform coupled with a plurality of physical fields in the foregoing first embodiment, the present application also provides an SGT MOS reliability analysis method coupled with a plurality of physical fields. Please refer to the accompanying drawings Figure 2 The SGT MOS reliability analysis method coupled with a plurality of physical fields includes the following steps:

[0088] The structural layers of the SGT MOS device are split to output a plurality of structural layers; damage state sample data of the plurality of structural layers under each physical field is imported, and independent influence coefficient matrices under a plurality of physical fields are constructed based on the damage state sample data; a coupled physical field is defined, wherein the coupled physical field is a physical field coupled by at least two physical fields; joint damage state sample data of the plurality of structural layers under the coupled physical field is imported, and coupled influence coefficient matrices under a plurality of coupled physical fields are constructed based on the joint damage state sample data; a reliability calculation module is constructed according to the independent influence coefficient matrices and the coupled influence coefficient matrices, and a first reliability calculation result of the SGT MOS device under a target coupled physical field is output according to the reliability calculation module.

[0089] Further, the plurality of structural layers at least include a metal gate layer, a polysilicon layer, a gate oxide layer, a channel region, and a substrate.

[0090] Further, the constructing the independent influence coefficient matrix under multiple physical fields based on the damage state sample data comprises: performing the same interval mapping on each index in the damage state sample data to output normalized damage score sample data; analyzing the normalized damage score sample data to calculate average damage score sample data of each structure layer under each physical field; evaluating the influence degree on the SGT MOS device according to the average damage score sample data of each structure layer under each physical field, and defining an independent influence coefficient of each structure layer under each physical field; and performing matrix arrangement on the independent influence coefficients under multiple physical fields to generate an independent influence coefficient matrix.

[0091] Further, the constructing the coupling influence coefficient matrix under multiple coupling physical fields based on the joint damage state sample data comprises: performing the same interval mapping on each index in the joint damage state sample data to output normalized joint damage score sample data; performing coupling calculation according to the normalized damage score sample data and the normalized joint damage score sample data to obtain a ratio of joint damage score sample data of each structure layer to average damage score sample data under a corresponding single physical field, and output a coupling enhancement factor of each structure layer under each physical field; defining a coupling influence coefficient vector based on the coupling enhancement factor to obtain a coupling influence coefficient matrix.

[0092] Further, the constructing the reliability calculation module comprises:

[0093] According to the independent influence coefficient matrix and the coupling influence coefficient matrix, a structure layer influence vector is constructed; according to proportions of the multiple structure layer historical failure frequencies in total failure events, a structure layer weight vector corresponding to the multiple structure layers is constructed; and the structure layer weight vector is used to perform weighted connection on the structure layer influence vector to construct a reliability calculation module.

[0094] Further, the reliability calculation module further comprises a neighborhood influence calculation module, wherein the neighborhood influence calculation module comprises a damage score judgment unit and a neighborhood diffusion calculation unit; the damage score judgment unit receives the normalized damage score sample data and the normalized joint damage score sample data to obtain a damage score sample data combination of each structure layer; according to the damage score sample data combination of each structure layer, an identified damage score sample data combination greater than a preset damage score threshold is screened; and the neighborhood diffusion calculation unit calculates a diffusion influence coefficient of each structure layer caused by neighborhood structure layer diffusion by using the identified damage score sample data combination to construct a neighborhood diffusion influence coefficient matrix.

[0095] Further, the outputting the first reliability calculation result of the SGT MOS device under a target coupling physical field according to the reliability calculation module comprises:

[0096] The neighborhood influence calculation module obtains a second reliability calculation result through the neighborhood diffusion influence coefficient matrix, and updates the first reliability calculation result according to the second reliability calculation result.

[0097] Further, the SGT MOS reliability analysis method based on multi-physical field coupling further comprises:

[0098] A three-dimensional multi-layer structure model of the SGT MOS device is constructed by connecting a finite element simulation system and a semiconductor multi-physical field simulation tool; a single physical field and a coupled physical field are loaded on the three-dimensional multi-layer structure model to obtain a plurality of structure layer simulation response data, and damage state sample data and joint damage state sample data are extracted from the plurality of structure layer simulation response data.

[0099] Further, the single physical field and the coupled physical field loaded on the three-dimensional multi-layer structure model are loaded by setting a nonlinear boundary condition; wherein the nonlinear boundary condition comprises one or more of an electric field loading boundary condition, a thermal field loading boundary condition, a stress field loading boundary condition, and an irradiation field loading boundary condition.

[0100] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The foregoing Figure 1 The multi-physical field coupled SGT MOS reliability analysis platform in Embodiment One and the specific examples are also applicable to the multi-physical field coupled SGT MOS reliability analysis method of the present embodiment. Through the foregoing detailed description of the multi-physical field coupled SGT MOS reliability analysis platform, those skilled in the art can clearly understand the multi-physical field coupled SGT MOS reliability analysis method in the present embodiment. Therefore, in order to make the specification concise, it will not be described in detail here.

[0101] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0102] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application also intends to include these modifications and variations.

Claims

1. A multi-physics coupled SGT MOS reliability analysis system, characterized in that, include: The structure layer splitting module is used to split the structure layer of the SGT MOS device and output multiple structure layers; The independent coefficient matrix construction module is used to import damage state sample data of the multiple structural layers under each physical field, and construct an independent influence coefficient matrix under multiple physical fields based on the damage state sample data. The coupled physical field definition module is used to define coupled physical fields, wherein the coupled physical field is a physical field generated by the coupling of at least two physical fields; The coupling coefficient matrix construction module is used to import the joint damage state sample data of the multiple structural layers under the coupled physical field, and construct the coupling influence coefficient matrix under the multiple coupled physical fields based on the joint damage state sample data. The reliability analysis module is used to construct a reliability calculation module based on the independent influence coefficient matrix and the coupling influence coefficient matrix, and output the first reliability calculation result of the SGT MOS device under the target coupled physical field based on the reliability calculation module.

2. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 1, characterized in that, The reliability analysis module also includes: An influence vector construction unit is used to construct a structural layer influence vector based on the independent influence coefficient matrix and the coupled influence coefficient matrix. The weight vector construction unit is used to construct the structural layer weight vectors corresponding to the multiple structural layers based on the proportion of the historical failure frequency of the multiple structural layers in the total failure events. The weighted connection unit is used to perform a weighted connection between the structural layer weight vector and the structural layer influence vector to construct a reliability calculation module.

3. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 1, characterized in that, The plurality of structural layers include at least a metal gate layer, a polysilicon layer, a gate oxide layer, a channel region, and a substrate.

4. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 1, characterized in that, The independent coefficient matrix construction module includes: The indicator mapping unit is used to map each indicator in the damage state sample data to the same interval and output normalized damage score sample data. The damage score calculation unit is used to analyze the normalized damage score sample data and calculate the average damage score sample data of each structural layer under each physical field. The influence coefficient determination unit is used to evaluate the degree of influence on the SGT MOS device based on the average damage score sample data of each structural layer under each physical field, and to define the independent influence coefficient of each structural layer under each physical field. The matrix arrangement unit is used to arrange the independent influence coefficients under multiple physical fields into a matrix to generate an independent influence coefficient matrix.

5. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 4, characterized in that, The coupling coefficient matrix construction module includes: The joint damage index mapping unit is used to map each index in the joint damage state sample data to the same interval and output normalized joint damage score sample data. The enhancement factor determination unit is used to perform coupled calculations based on the normalized damage score sample data and the normalized joint damage score sample data to obtain the ratio of the joint damage score sample data of each structural layer to the average damage score sample data under the action of the corresponding individual physical field, and output the coupling enhancement factor of each structural layer under each physical field. The coupling matrix determination unit is used to define the coupling influence coefficient vector based on the coupling enhancement factor to obtain the coupling influence coefficient matrix.

6. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 5, characterized in that, The reliability calculation module further includes a neighborhood influence calculation module, wherein the neighborhood influence calculation module includes a damage scoring judgment unit and a neighborhood diffusion calculation unit; The damage scoring judgment unit receives the normalized damage scoring sample data and the normalized joint damage scoring sample data to obtain a combination of damage scoring sample data for each structural layer. Based on the combination of damage score sample data for each structural layer, select the combination of identification damage score sample data that exceeds the preset damage score threshold. The neighborhood diffusion calculation unit uses the identified damage score sample data to calculate the diffusion influence coefficient of each structural layer on the diffusion of the neighboring structural layer, and constructs the neighborhood diffusion influence coefficient matrix.

7. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 6, characterized in that, The reliability analysis module also includes: The second reliability calculation unit is used by the neighborhood influence calculation module to obtain the second reliability calculation result through the neighborhood diffusion influence coefficient matrix. The calculation result update unit is used to update the first reliability calculation result according to the second reliability calculation result.

8. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 1, characterized in that, The multiphysics coupled SGT MOS reliability analysis platform also includes: The 3D model building module is used to connect the finite element simulation system and the semiconductor multiphysics simulation tool to build a 3D multilayer structure model of the SGTMOS device; The damage analysis module is used to load a single physical field and coupled physical fields onto the three-dimensional multi-layer structure model, obtain simulated response data of multiple structural layers, and extract damage state sample data and joint damage state sample data from the simulated response data of multiple structural layers.

9. The SGT MOS reliability analysis system with multi-physics coupling as described in claim 8, characterized in that, The three-dimensional multi-layer structure model is loaded with single and coupled physical fields by setting nonlinear boundary conditions; The nonlinear boundary conditions include one or more of the following: electric field loading boundary conditions, thermal field loading boundary conditions, stress field loading boundary conditions, and irradiation field loading boundary conditions.

10. A reliability analysis method for SGT MOS with multi-physics coupling, characterized in that, The reliability analysis method for SGT MOS, which is performed by the multi-physics coupled SGT MOS reliability analysis system according to any one of claims 1 to 9, comprises: The SGT MOS device is decomposed into multiple structural layers. Import damage state sample data of the multiple structural layers under each physical field, and construct an independent influence coefficient matrix under multiple physical fields based on the damage state sample data; Define a coupled physical field, wherein the coupled physical field is a physical field generated by the coupling of at least two physical fields; Import the joint damage state sample data of the multiple structural layers under the coupled physical field, and construct the coupling influence coefficient matrix under the multiple coupled physical fields based on the joint damage state sample data; A reliability calculation module is constructed based on the independent influence coefficient matrix and the coupling influence coefficient matrix, and the first reliability calculation result of the SGT MOS device under the target coupled physical field is output based on the reliability calculation module.

Images