Software tool and method for analyzing and predicting electromigration risk

Abstract

The embodiment of the invention provides a software tool and method for analyzing and predicting an electromigration risk, and the software tool comprises an input interface which is used for importing integrated circuit design parameters; the integrated circuit design parameters at least comprise circuit layout, wire material, current density, working temperature and stress state; the electromigration simulation module is used for carrying out simulation analysis on the electromigration risk in the integrated circuit design based on a preset algorithm and predicting the electromigration risk existing in the integrated circuit design; the report generation module is used for generating an electromigration risk report; and the optimization suggestion module is used for generating a suggestion for optimizing the design of the integrated circuit according to the electromigration risk report. According to the embodiment of the invention, effective analysis and prediction of the electromigration risk are realized.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of integrated circuit technology, and specifically to a software tool and method for analyzing and predicting electromigration risks. Background Art

[0002] In the field of integrated circuit (IC) design, electromigration is a critical issue that cannot be ignored, especially in high-density and high-power applications, where its impact is even more prominent. Electromigration is a phenomenon in which metal atoms migrate due to the flow of electrons. Over time, this phenomenon can cause metal wires to deform, create voids, or even break, leading to circuit failure. In modern microelectronic devices, as device size continues to shrink and power density continues to increase, the current density within the wires has increased significantly, significantly increasing the risk of electromigration.

[0003] In this context, how to provide a software tool for analyzing and predicting electromigration risks has become an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, embodiments of the present application provide a software tool and method for analyzing and predicting electromigration risks, so as to analyze and predict electromigration risks.

[0005] To achieve the above objectives, the embodiments of the present application provide the following technical solutions.

[0006] In a first aspect, embodiments of the present application provide a software tool for analyzing and predicting electromigration risks, including:

[0007] An input interface for importing integrated circuit design parameters; the integrated circuit design parameters include at least circuit layout, wire material, current density, operating temperature, and stress state;

[0008] An electromigration simulation module is used to simulate and analyze electromigration risks in integrated circuit design based on a preset algorithm, and to predict the electromigration risks in integrated circuit design;

[0009] Report generation module, used to generate electromigration risk report;

[0010] An optimization suggestion module is used to generate suggestions for optimizing the integrated circuit design based on the electromigration risk report.

[0011] Optionally, the electromigration simulation module includes:

[0012] Multi-physics simulation module, which is used to establish an electromigration stress model of interconnects based on current density, operating temperature, and stress state based on finite element analysis to locate areas with electromigration risk;

[0013] a failure life prediction module, configured to calculate the expected failure time of each interconnected line segment based on the first formula, and to screen out areas whose lifespan is lower than a preset lifespan;

[0014] A machine learning recognition module, which uses support vector machine and random forest algorithms to train classifiers based on simulation data to automatically identify areas with electromigration risk in different topologies.

[0015] The simulation iteration module is used to perform multiple simulation iterations based on Monte Carlo simulation technology to evaluate the probability distribution of electromigration failure.

[0016] Optionally, the electromigration simulation module further includes:

[0017] Adaptive mesh refinement module, used to automatically refine the simulation mesh in areas where the electromigration risk is greater than a preset risk value;

[0018] The local feature analysis module is used to extract features and perform sensitivity analysis on areas where the current density is greater than the preset current density and where the temperature gradient changes sharply.

[0019] Optionally, the integrated circuit design parameters are imported into an input interface via a graphical interface or a text file.

[0020] Optionally, the electromigration risk report includes locations where electromigration risk exists in the integrated circuit design, and risk levels of the electromigration risk at the locations where electromigration risk exists.

[0021] Optionally, also include:

[0022] Visualization tools for graphically displaying electromigration risk analysis results;

[0023] The visualization tool includes multiple visualization options.

[0024] Optionally, the suggestion for optimizing the integrated circuit design includes:

[0025] Recommendations for wiring optimization, material selection, and process parameter adjustment.

[0026] Optionally, the first formula is:

[0027]

[0028] Where MTTF is the mean time to failure, A is a constant, j is the current density, j0 is the reference current density, n is the electromigration index, Ea is the atomic migration activation energy, k is the Boltzmann constant, and T is the absolute temperature.

[0029] In a second aspect, an embodiment of the present application provides a method for analyzing and predicting electromigration risk, which is implemented based on the software tool described in the first aspect, including:

[0030] importing integrated circuit design parameters through an input interface in the software tool; the integrated circuit design parameters include at least circuit layout, conductor material, current density, operating temperature, and stress state;

[0031] Using the electromigration simulation module in the software tool, simulate and analyze the electromigration risk in the integrated circuit design based on the imported integrated circuit design parameters, and predict the electromigration risk in the integrated circuit design;

[0032] Obtaining an electromigration risk report generated by a report generation module in the software tool;

[0033] Modifying the integrated circuit design based on the suggestions for optimizing the integrated circuit design generated by the optimization suggestion module in the software tool.

[0034] Optionally, also include:

[0035] Obtaining graphical electromigration risk analysis results displayed by a visualization tool in the software tool.

[0036] Compared with the prior art, the technical solution of the embodiment of the present application has the following advantages:

[0037] An embodiment of the present application provides a software tool and method for analyzing and predicting electromigration risks, wherein the software tool includes: an input interface for importing integrated circuit design parameters; the integrated circuit design parameters include at least circuit layout, wire material, current density, operating temperature and stress state; an electromigration simulation module for simulating and analyzing the electromigration risks in the integrated circuit design based on a preset algorithm, and predicting the electromigration risks existing in the integrated circuit design; a report generation module for generating an electromigration risk report; and an optimization suggestion module for generating suggestions for optimizing the integrated circuit design based on the electromigration risk report.

[0038] As can be seen, the software tool provided in the embodiments of this application imports integrated circuit design parameters including circuit layout, wire material, current density, operating temperature, and stress state through an input interface, providing a data foundation for subsequent analysis. The electromigration simulation module then uses a preset algorithm to simulate and analyze the electromigration risk in the integrated circuit design, predicting the potential electromigration risk in the integrated circuit design, thereby achieving effective analysis and prediction of electromigration risk.

[0039] Furthermore, the report generation module of the present application generates an electromigration risk report, so that the optimization suggestion module generates suggestions for optimizing the integrated circuit design based on the electromigration risk report, thereby enabling users to foresee and resolve electromigration risks during the design phase and improve design efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0041] Figure 1 This is a schematic diagram of an optional structure of a software tool for analyzing and predicting electromigration risks provided in an embodiment of the present application;

[0042] Figure 2 This is a schematic diagram of an optional structure of another software tool for analyzing and predicting electromigration risks provided in an embodiment of the present application;

[0043] Figure 3 This is an optional flow chart of a method for analyzing and predicting electromigration risks provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0045] As described in the background technology, in the field of integrated circuit (IC) design, electromigration is a key issue that cannot be ignored, especially in high-density and high-power application scenarios, its impact is more prominent. Electromigration is a phenomenon of metal atom migration caused by the flow of electrons. Under the action of a long time, this phenomenon can cause the metal wire to deform, produce voids or even break, thereby causing circuit failure. In modern microelectronic devices, as device size continues to shrink and power density continues to increase, the current density in the wire rises significantly, significantly increasing the risk of electromigration. In this context, how to provide a software tool for analyzing and predicting electromigration risk, to analyze and predict electromigration risk, has become a problem urgently to be solved by those skilled in the art.

[0046] The inventors analyzed and found that the current electromigration risk assessment mainly relies on engineers' manual experience and static current density rules for auxiliary judgment. For example, by simply applying empirical thresholds (such as the current density upper limit of 1MA / cm2) to determine whether the interconnect line has electromigration risk.

[0047] However, this method usually only uses current density as a reference and fails to comprehensively consider complex factors such as temperature gradients, mechanical stress, and material micro-defects, resulting in one-sided judgments. Moreover, for local hotspots in complex structures such as multi-layer interconnects and via arrays, manual experience is difficult to accurately predict. Furthermore, this method can only provide vague risk warnings and cannot accurately give failure probability or life prediction values. In addition, different engineers make different judgments based on their own experience, resulting in poor consistency.

[0048] Due to the above defects, the accuracy of traditional empirical judgment methods in practical applications is low, usually only achieving a risk identification accuracy of about 60%-70%. When facing advanced process nodes (such as 7nm and 5nm), the accuracy drops further.

[0049] In view of this, an embodiment of the present application provides a software tool and method for analyzing and predicting electromigration risks, wherein the software tool includes: an input interface for importing integrated circuit design parameters; the integrated circuit design parameters include at least circuit layout, wire material, current density, operating temperature and stress state; an electromigration simulation module for simulating and analyzing the electromigration risks in the integrated circuit design based on a preset algorithm, and predicting the electromigration risks existing in the integrated circuit design; a report generation module for generating an electromigration risk report; and an optimization suggestion module for generating suggestions for optimizing the integrated circuit design based on the electromigration risk report.

[0050] As can be seen, the software tool provided in the embodiments of this application imports integrated circuit design parameters including circuit layout, wire material, current density, operating temperature, and stress state through an input interface, providing a data foundation for subsequent analysis. The electromigration simulation module then uses a preset algorithm to simulate and analyze the electromigration risk in the integrated circuit design, predicting the potential electromigration risk in the integrated circuit design, thereby achieving effective analysis and prediction of electromigration risk.

[0051] Furthermore, the report generation module of the present application generates an electromigration risk report, so that the optimization suggestion module generates suggestions for optimizing the integrated circuit design based on the electromigration risk report, thereby enabling users to foresee and resolve electromigration risks during the design phase and improve design efficiency.

[0052] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0053] refer to Figure 1 An optional structural diagram of a software tool for analyzing and predicting electromigration risks is shown as an example. Figure 1 As shown, the software tool includes an input interface 10 , an electromigration simulation module 20 , a report generation module 30 and an optimization suggestion module 40 .

[0054] Input interface 10 is used to import integrated circuit design parameters; these integrated circuit design parameters include at least circuit layout, conductor material, current density, operating temperature, and stress state. In an optional implementation, these integrated circuit design parameters can be imported into input interface 10 via a graphical interface or text file. In this embodiment, input interface 10 supports integrated circuit design files in multiple formats, including but not limited to standard formats such as GDSII, LEF / DEF, and Verilog, to facilitate user parameter import, reduce input data preparation time, and reduce the possibility of input errors.

[0055] The electromigration simulation module 20 is used to simulate and analyze the electromigration risk in the integrated circuit design based on a preset algorithm, and predict the electromigration risk in the integrated circuit design. In an optional implementation, refer to Figure 2 Another optional structural diagram of a software tool for analyzing and predicting electromigration risks is shown as an example. Figure 2 As shown, the electromigration simulation module 20 includes: a multi-physics field simulation module 21, a failure life prediction module 22, a machine learning identification module 23 and a simulation iteration module 24.

[0056] Among them, the multi-physics field simulation module 21 is used to establish an electromigration stress model of the interconnection wires based on the finite element analysis method, which is used to locate the areas where there is electromigration risk. The multi-physics field simulation module 21 uses the finite element analysis method based on the imported integrated circuit design parameters to establish an electromigration stress model of the interconnection wires based on the current density, operating temperature and stress state. The multi-physics field simulation module simultaneously considers the coupling effects of the electric field, temperature field and stress field, and calculates the current density distribution, temperature distribution and stress distribution of each point by solving a set of partial differential equations. During the calculation process, the multi-physics field simulation module takes into account the nonlinear changes of material properties, such as the change of resistivity with temperature, the change of material elastic modulus with temperature, etc. Through these calculations, the multi-physics field simulation module can identify areas with concentrated current density, areas with large temperature gradients and areas with concentrated stress, which are usually areas with higher electromigration risks.

[0057] The failure life prediction module 22 is used to calculate the expected failure time of each interconnected line segment based on the first formula and filter out areas with a lifespan lower than a preset lifespan. The first formula is: Where MTTF is mean time to failure, A is a constant, j is current density, j0 is reference current density, n is electromigration index, Ea is atomic migration activation energy, k is Boltzmann constant, and T is absolute temperature. In this embodiment, the failure life prediction module can set different parameter values ​​according to different wire materials. For copper interconnects, the value range of A is 1.0×10 8 to 5.0×10 8 , n ranges from 1.0 to 2.0, Ea ranges from 0.7eV to 1.0eV; for aluminum interconnects, A ranges from 5.0×10 7 to 2.0×10 8 , n ranges from 1.5 to 2.5, and Ea ranges from 0.5 eV to 0.8 eV. The reference current density j0 can be, for example, 1.0×10 6 A / cm 2 The failure life prediction module compares the calculated MTTF value with a preset life threshold (e.g., 10 years or 87,600 hours), screens out areas with a lifespan lower than the preset lifespan, and marks these areas as high-risk areas.

[0058] The machine learning identification module 23 is used to use the support vector machine algorithm and the random forest algorithm to train the classifier based on simulation data to automatically identify areas with electromigration risks in different topological structures. The machine learning identification module 23 first extracts features from historical simulation data and actual failure cases, including current density, temperature gradient, stress gradient, wire width, wire length, corner angle, etc. These features are then used to train the support vector machine classifier and the random forest classifier. The support vector machine classifier is mainly used for binary classification problems, that is, to determine whether a certain area has electromigration risks; the random forest classifier is used for multi-classification problems, that is, to classify the risk level into three levels: low, medium, and high. The machine learning identification module uses a cross-validation method to evaluate the classifier performance to ensure that the classification accuracy reaches the preset accuracy. In practical applications, the machine learning identification module can identify potential risk areas that are difficult to detect with traditional methods, especially risk areas in complex topological structures, such as via arrays, narrow wiring areas, etc.

[0059] The simulation iteration module 24 is used to perform multiple simulation iterations based on Monte Carlo simulation technology to evaluate the probability distribution of electromigration failure. The simulation iteration module 24 takes into account parameter fluctuations in the manufacturing process, such as line width fluctuations, thickness fluctuations, material property fluctuations, etc. In each iteration, the simulation iteration module 24 randomly generates parameter values ​​that conform to a specific distribution, and then calls the multi-physics simulation module and the failure life prediction module to perform calculations. Normally, the simulation iteration module will perform 1,000 to 10,000 iterations to ensure the reliability of the statistical results. By analyzing these iterative results, the simulation iteration module 24 can derive the probability distribution of electromigration failure, including statistical characteristics such as the mean value, standard deviation, minimum value, and maximum value of the failure time. These statistical results can help designers understand the electromigration risk more comprehensively, especially in risk assessment considering manufacturing fluctuations.

[0060] In an optional implementation, the electromigration simulation module may further include: an adaptive grid refinement module 25 and a local feature analysis module 26 .

[0061] Among them, the adaptive grid refinement module 25 is used to automatically refine the simulation grid in areas where the electromigration risk is greater than a preset risk value. The adaptive grid refinement module identifies areas with large current density, temperature gradient or stress gradient based on the preliminary simulation results, and then automatically increases the grid density in these areas. In an optional implementation, the grid refinement can use a quadtree or octree algorithm, so that the calculation accuracy of key areas can be improved while maintaining calculation efficiency. In this embodiment, when the electromigration risk value of a certain area exceeds a preset threshold, the adaptive grid refinement module will reduce the grid size of the area to 1 / 4 to 1 / 8 of the original size.

[0062] The local feature analysis module 26 is used to extract features and perform sensitivity analysis on areas where the current density is greater than a preset current density and where the temperature gradient changes sharply. The local feature analysis module 26 first identifies areas where the current density exceeds a preset threshold or the temperature gradient exceeds a preset threshold, and then performs a detailed analysis of these areas. The local feature analysis module extracts the geometric features of these areas, such as line width, line length, corner angle, cross structure, etc., and analyzes the correlation between these features and electromigration risk. At the same time, the local feature analysis module also performs parameter sensitivity analysis, and by changing a certain parameter and observing the risk change, determines which parameters have the greatest impact on electromigration risk. These analysis results provide an important basis for subsequent design optimization.

[0063] In a specific implementation, the electromigration simulation module 20 in the software tool performs electromigration risk simulation based on imported integrated circuit design parameters (including layout data, current load conditions, ambient temperature, etc.). During the simulation process, the current density distribution, local temperature field changes, and material physical properties (such as diffusion coefficient and activation energy) are comprehensively considered. The module also identifies potential high-risk areas in the integrated circuit by solving the following electromigration-related mathematical model:

[0064] Electromigration Mass Transport Equation:

[0065] Where C is the vacancy concentration, D is the diffusion coefficient, and Z * is the effective charge number, e is the electron charge, ρ is the resistivity, j is the current density, k is the Boltzmann constant, and T is the absolute temperature. The left side of the equation represents the rate of change of the vacancy concentration with time, while the right side represents the divergence of the vacancy flux due to the combined effects of diffusion and electromigration. When the right side of the equation is positive, it indicates a net outflow of vacancies from the point; when the right side is negative, it indicates a net inflow of vacancies. By solving this equation, we can obtain the spatial and temporal distribution of the vacancy concentration under given initial and boundary conditions.

[0066] The expected failure time of each interconnected segment is calculated based on the first formula, which is a modified Black's equation. The specific formula is as follows:

[0067] Where MTTF is the mean time to failure, A is a constant, j is the current density, j0 is the reference current density, n is the electromigration index, Ea is the atomic migration activation energy, k is the Boltzmann constant, and T is the absolute temperature.

[0068] Thermal migration effect correction model: Considering the diffusion effect caused by temperature gradient (Thermal Gradient Effect, TGE), the influence of temperature change on vacancy migration velocity is reflected by correcting the diffusion flow term.

[0069] During the simulation process, the software tool first calculates the steady-state temperature and current density distributions across the entire layout using finite element analysis. Using these as boundary conditions, it iteratively solves the electromigration mass transport equation, dynamically simulating the atomic migration process and vacancy accumulation within the material. Then, combined with a modified Black's equation, it generates a failure probability assessment and lifetime prediction for each interconnect segment, identifying areas with a high risk of electromigration failure.

[0070] The report generation module 30 is used to generate an electromigration risk report. The electromigration risk report includes the locations where electromigration risks exist in the integrated circuit design, and the risk levels of the electromigration risks at the locations where electromigration risks exist. The report generation module integrates the analysis results of the electromigration simulation module into a structured report, including risk overview, detailed risk list, risk statistical analysis and other parts. The risk overview part provides the overall risk assessment results, such as the number of high-risk areas, the number of medium-risk areas, the number of low-risk areas, etc. The detailed risk list part lists the specific information of each risk area, including location coordinates, risk level, expected failure time, failure probability, etc. The risk statistical analysis part provides statistical characteristics of risk distribution, such as risk level distribution, failure time distribution, etc. The report generation module supports multiple report formats, which may include but are not limited to HTML, PDF, CSV, etc., so that users can choose the appropriate format according to their needs.

[0071] The optimization suggestion module 40 is used to generate suggestions for optimizing the integrated circuit design based on the electromigration risk report. The suggestions for optimizing the integrated circuit design may include: wiring optimization suggestions, material selection suggestions, and process parameter adjustment suggestions. Wiring optimization suggestions are mainly aimed at areas where the current density is too high or the stress is concentrated, and optimization solutions such as changing the wire width, adjusting the wire path, and adding parallel wires are proposed. Material selection suggestions are mainly aimed at specific application scenarios, and more suitable wire material selections are proposed, such as using more high-temperature resistant materials in high-temperature environments, and using materials with lower resistivity in high-current density environments. Process parameter adjustment suggestions are mainly aimed at the manufacturing process, and suggestions are proposed to adjust process parameters such as deposition temperature, annealing time, and etching parameters to improve the reliability of interconnects.

[0072] In an optional implementation, the software tool may further include: a visualization tool for graphically displaying the analysis results of electromigration risks; wherein the visualization tool includes a variety of visualization options. Specifically, the visualization tool supports a variety of visualization options, including two-dimensional plan views, three-dimensional stereo views, heat maps, contour maps, etc. The two-dimensional plan view is suitable for viewing the overall risk distribution; the three-dimensional stereo view is suitable for viewing the relationship between risk and physical structure; the heat map is suitable for viewing the current density or temperature distribution; and the contour map is suitable for viewing the stress distribution. In an optional implementation, the visualization tool may also support interactive operations, such as zooming, panning, and rotating, to facilitate users to view the risk distribution from different angles and scales.

[0073] In an optional implementation, during the process of analyzing the electromigration characteristics by the electromigration simulation module 20 , different machine learning algorithms and efficient finite element analysis methods may be used to further improve the calculation speed and recognition accuracy.

[0074] Among them, different machine learning algorithms can include:

[0075] Gradient Boosting Decision Tree (GBDT): Gradually improves prediction accuracy by building multiple weak classifiers (decision trees), and is suitable for processing complex data involving multiple interactions among electromigration factors.

[0076] Extreme Gradient Boosting (XGBoost): An optimized version of GBDT with faster training speed and higher accuracy, suitable for risk point identification in large-scale simulation datasets.

[0077] Convolutional Neural Network (CNN): Used to directly extract local structural features based on layout images, thereby realizing automated high-risk area identification.

[0078] Graph Neural Network (GNN): It is particularly suitable for modeling the structure of interconnected networks (layoutnetlists) and can learn electromigration risk characteristics through the relationship between nodes and edges.

[0079] Self-supervised learning method: In the absence of labeled data, the model is trained through generative tasks to improve the migration prediction ability on new layouts.

[0080] Efficient finite element analysis algorithms can include:

[0081] Multigrid Method: By accelerating convergence on grids of different scales, it significantly reduces the computational time for solving large-scale finite element problems.

[0082] Adaptive Weighted Residual Method: Dynamically refines or coarsens the grid based on local errors, optimizes computing resource allocation, and improves solution efficiency.

[0083] Sparse Direct Solvers (such as PARDISO): They provide fast and efficient solutions for sparse linear systems and are suitable for large-scale electromigration coupled simulations.

[0084] Parallel Finite Element Analysis: Based on multi-core CPU or GPU acceleration, distributed solution of electromigration models improves the simulation speed of large layouts.

[0085] The software tools provided in the embodiments of this application are suitable for various high-density and high-power integrated circuit designs, including but not limited to microprocessors, high-performance computing chips, communication chips, etc., and have broad application prospects. Furthermore, the embodiments of this application, through multi-physics field simulation and the combination of various algorithms, can increase the accuracy of electromigration risk identification to over 90%, significantly improving the accuracy of electromigration risk identification and solving the problem that traditional empirical judgment methods have low response to electromigration fault points and cannot accurately predict electromigration risks from the design stage.

[0086] As can be seen, the software tool provided in the embodiments of this application imports integrated circuit design parameters including circuit layout, wire material, current density, operating temperature, and stress state through an input interface, providing a data foundation for subsequent analysis. The electromigration simulation module then uses a preset algorithm to simulate and analyze the electromigration risk in the integrated circuit design, predicting the potential electromigration risk in the integrated circuit design, thereby achieving effective analysis and prediction of electromigration risk.

[0087] Furthermore, the report generation module of the present application generates an electromigration risk report, so that the optimization suggestion module generates suggestions for optimizing the integrated circuit design based on the electromigration risk report, thereby enabling users to foresee and resolve electromigration risks during the design phase and improve design efficiency.

[0088] In a further implementation, the present application also provides a method for analyzing and predicting electromigration risk, which can be implemented based on the software tool described above. Figure 3 An optional flow chart of a method for analyzing and predicting electromigration risk provided by an embodiment of the present application is shown. The method may include:

[0089] Step S100: importing integrated circuit design parameters through an input interface in the software tool.

[0090] The integrated circuit design parameters include at least circuit layout, conductor material, current density, operating temperature and stress state. In this embodiment, the user can import the integrated circuit design parameters into the input interface through a graphical interface or a text file.

[0091] Step S200 , utilizing the electromigration simulation module in the software tool to simulate and analyze the electromigration risk in the integrated circuit design according to the imported integrated circuit design parameters, and predicting the electromigration risk in the integrated circuit design.

[0092] In this embodiment, the electromigration simulation process is divided into several stages. First, the multi-physics field simulation module establishes an electromigration stress model of the interconnection line based on current density, operating temperature and stress state. The model takes into account factors such as momentum transfer caused by current, thermal migration caused by temperature gradient, and stress migration caused by stress gradient. Then, the failure life prediction module calculates the expected failure time of each interconnection segment based on the first formula. Next, the machine learning identification module uses the support vector machine algorithm and the random forest algorithm to identify areas with electromigration risks in different topological structures. Finally, the simulation iteration module performs multiple simulation iterations based on Monte Carlo simulation technology to evaluate the probability distribution of electromigration failure. During the simulation process, the adaptive grid refinement module refines the grid in areas where the risk value exceeds the preset threshold, and the grid size is reduced to 1 / 4 to 1 / 8 of the original; the local feature analysis module performs feature extraction and sensitivity analysis on areas where the current density is greater than the preset current density and where the temperature gradient changes sharply.

[0093] Step S300: obtaining an electromigration risk report generated by a report generation module in the software tool.

[0094] Obtain an electromigration risk report generated by a report generation module in a software tool. The electromigration risk report includes the locations where electromigration risks exist in the integrated circuit design, and the risk levels of the electromigration risks at the locations where electromigration risks exist. In this embodiment, the risk levels can be divided into three levels: high risk (expected life less than 5 years), medium risk (expected life between 5 and 10 years) and low risk (expected life greater than 10 years but less than 15 years). The electromigration risk report may include a risk overview, a detailed risk list and a risk statistical analysis. The risk overview shows how many high-risk areas, medium-risk areas and low-risk areas there are in the entire design, and the proportion of these areas in the entire design. The detailed risk list lists the specific information of each risk area, including location coordinates, risk level, expected failure time, failure probability, etc. The risk statistical analysis provides statistical charts of risk distribution, such as risk level pie charts, failure time histograms, etc.

[0095] Step S400 , modifying the integrated circuit design according to the suggestions for optimizing the integrated circuit design generated by the optimization suggestion module in the software tool.

[0096] In this embodiment, the optimization suggestions include wiring optimization suggestions, material selection suggestions, and process parameter adjustment suggestions. The wiring optimization suggestions propose specific wiring modification plans for high-risk areas, such as widening a wire with a width of 0.5μm to 0.8μm, or dividing a single wire into two parallel wires to reduce the current density. The material selection suggestions propose more suitable material choices for specific application scenarios, such as using Cu-Mn alloy instead of pure copper in a high-temperature environment, or using Cu-Al alloy instead of aluminum in a high current density environment. The process parameter adjustment suggestions propose specific parameter adjustment plans for the manufacturing process, such as increasing the copper deposition temperature from 250°C to 300°C to improve the grain structure, or extending the annealing time from 30 minutes to 45 minutes to reduce residual stress. The user can modify the integrated circuit design based on these suggestions, and then can re-import the modified design parameters for verification to ensure that the modified design meets the reliability requirements.

[0097] In an optional implementation, the method may further include: obtaining graphical analysis results of electromigration risks displayed by a visualization tool in the software tool. The visualization tool includes a variety of visualization options. Specifically, the visualization tool supports a variety of visualization options, including two-dimensional plane views, three-dimensional stereo views, heat maps, contour maps, etc. The two-dimensional plane view is suitable for viewing the overall risk distribution; the three-dimensional stereo view is suitable for viewing the relationship between risk and physical structure; the heat map is suitable for viewing the current density or temperature distribution; the contour map is suitable for viewing the stress distribution. In an optional implementation, the visualization tool may also support interactive operations, such as zooming, panning, rotating, etc., to facilitate users to view the risk distribution from different angles and scales.

[0098] The above describes multiple embodiment schemes provided by the embodiments of the present application. The various optional methods introduced in each embodiment scheme can be combined and cross-referenced with each other without conflict, thereby extending a variety of possible embodiment schemes, which can all be considered as embodiment schemes disclosed and open in the embodiments of the present application.

[0099] Although the embodiments of the present application are disclosed above, the present application is not limited thereto. Any person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims.

Claims

1. A software tool for analyzing and predicting electromigration risk, characterized in that: include: Input interface, used to import integrated circuit design parameters; The integrated circuit design parameters include at least circuit layout, conductor material, current density, operating temperature and stress state; An electromigration simulation module is used to simulate and analyze electromigration risks in integrated circuit design based on a preset algorithm, and to predict the electromigration risks in integrated circuit design; Report generation module, used to generate electromigration risk report; An optimization suggestion module is used to generate suggestions for optimizing the integrated circuit design based on the electromigration risk report.

2. The software tool according to claim 1, wherein: The electromigration simulation module includes: Multi-physics simulation module, which is used to establish an electromigration stress model of interconnects based on current density, operating temperature, and stress state based on finite element analysis to locate areas with electromigration risk; a failure life prediction module, configured to calculate the expected failure time of each interconnected line segment based on the first formula, and to screen out areas whose lifespan is lower than a preset lifespan; A machine learning recognition module, which uses support vector machine and random forest algorithms to train classifiers based on simulation data to automatically identify areas with electromigration risk in different topologies. The simulation iteration module is used to perform multiple simulation iterations based on Monte Carlo simulation technology to evaluate the probability distribution of electromigration failure.

3. The software tool according to claim 2, wherein: The electromigration simulation module further includes: Adaptive mesh refinement module, used to automatically refine the simulation mesh in areas where the electromigration risk is greater than a preset risk value; The local feature analysis module is used to extract features and perform sensitivity analysis on areas where the current density is greater than the preset current density and where the temperature gradient changes sharply.

4. The software tool according to claim 1, wherein: The integrated circuit design parameters are imported into an input interface via a graphical interface or a text file.

5. The software tool according to claim 1, wherein: The electromigration risk report includes locations where electromigration risk exists in the integrated circuit design, and risk levels of the electromigration risk at the locations where electromigration risk exists.

6. The software tool according to claim 1, wherein: Also includes: Visualization tools for graphically displaying electromigration risk analysis results; The visualization tool includes multiple visualization options.

7. The software tool according to claim 1, wherein: The suggestion for optimizing the integrated circuit design includes: Recommendations for wiring optimization, material selection, and process parameter adjustment.

8. The software tool according to claim 2, wherein: The first formula is: Where MTTF is the mean time to failure, A is a constant, j is the current density, j0 is the reference current density, n is the electromigration index, Ea is the atomic migration activation energy, k is the Boltzmann constant, and T is the absolute temperature.

9. A method for analyzing and predicting electromigration risk, implemented based on the software tool according to any one of claims 1 to 8, characterized in that: include: importing integrated circuit design parameters through an input interface in the software tool; The integrated circuit design parameters include at least circuit layout, conductor material, current density, operating temperature and stress state; Using the electromigration simulation module in the software tool, simulate and analyze the electromigration risk in the integrated circuit design based on the imported integrated circuit design parameters, and predict the electromigration risk in the integrated circuit design; Obtaining an electromigration risk report generated by a report generation module in the software tool; Modifying the integrated circuit design based on suggestions for optimizing the integrated circuit design generated by an optimization suggestion module in the software tool.

10. The method according to claim 9, characterized in that Also includes: Obtaining graphical electromigration risk analysis results displayed by a visualization tool in the software tool.

Images