A method and system for evaluating a multi-layer chip stacking solution

Abstract

The present invention relates to the technical field of stacking packaging process optimization, and in particular to a method and system for evaluating a multi-layer chip stacking scheme. The method comprises: acquiring several variable information sets by performing data collection on each determined scheme variable item, selecting process information for each variable scheme, acquiring several basic process schemes, performing directional adjustment on the basic process schemes, acquiring pre-selected stacking schemes, performing process digital simulation on each stacking scheme, and acquiring the best stacking scheme based on the simulation results. The present invention effectively solves the problem of low process debugging efficiency caused by multi-variable coupling in the stacking packaging process. By performing data collection and compatibility evaluation on multiple scheme variable items, the scheme can comprehensively analyze the mutual influence of various process parameters in different stacking schemes, thereby more accurately selecting the most suitable stacking scheme and improving production efficiency and product quality.

Description

Technical Field

[0001] The present invention relates to the technical field of stacked packaging process optimization, and in particular to an evaluation method and system for a multi-layer chip stacking solution. Background Art

[0002] As semiconductor packaging technology rapidly advances toward higher density and higher integration, stacked packaging, a key technology for achieving three-dimensional integration and miniaturized packaging, has become a core research area in the advanced packaging field. By stacking and interconnecting multiple chips, stacked packaging significantly improves device performance and reduces package size, finding widespread application in mobile devices, high-performance computing, and the Internet of Things. Driven by emerging demands such as 5G communications and artificial intelligence chips, stacked packaging is placing higher demands on the precision, reliability, and efficiency of stacking solutions, prompting the industry to explore intelligent process optimization methods that combine data-driven approaches with virtual simulation.

[0003] Existing stacked packaging processes often rely on experience or a single parameter optimization method when selecting a stacking solution, and are unable to fully consider the impact of all variable factors. Traditional stacking solution selection methods are usually based on historical data and empirical formulas, but these methods lack comprehensive digital simulation and precise tuning of the process, making it difficult to cope with complex packaging requirements and changing production environments. At the same time, there is also a lack of systematic evaluation of the compatibility and mutual influence between various process parameters. In actual applications, there is often uncertainty in the selection of solution variables and process adjustments, which not only reduces process efficiency but may also lead to unstable product quality.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0005] The present invention provides a method and system for evaluating a multi-layer chip stacking solution, which can effectively solve the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A method for evaluating a multi-layer chip stacking solution, the method comprising:

[0008] Determine solution variable items, and collect data for each solution variable item to obtain a plurality of variable information sets, wherein the solution variable items constitute a stacking solution in a stacking packaging process;

[0009] Each of the scheme variable items selects process information corresponding to the variable information set and obtains several basic process schemes;

[0010] Directively adjusting the process information of the solution variable items in the basic process solution to obtain a plurality of optimized process solutions, and using the plurality of optimized process solutions and the basic process solution as pre-selected lamination solutions;

[0011] Construct a virtual process mirror, perform digital process simulation on the pre-selected stacking schemes, and obtain stacking scheme feedback;

[0012] The corresponding preselected lamination scheme is selected as the optimal lamination scheme according to the lamination scheme feedback.

[0013] Furthermore, each of the scheme variable items selects the process information corresponding to the variable information set and obtains several basic process schemes, including:

[0014] Establishing a lamination process database, extracting historical lamination schemes according to the lamination process database, and marking the scheme variable items corresponding to the historical lamination schemes as pre-selected combination schemes, wherein the scheme variable items include glue type information, dispensing path information, and lamination geometry information;

[0015] Extracting a plurality of dispensing path information and lamination geometry information according to the lamination process database, and performing compatibility evaluation on each dispensing path information and a plurality of lamination geometry information;

[0016] Classifying the historical lamination schemes according to the glue type information to obtain a plurality of historical lamination scheme sets, wherein each of the historical lamination scheme sets corresponds to each of the glue type information;

[0017] The calling weight is set according to the compatibility evaluation result and the pre-selected combination scheme, and the process information of the remaining scheme variable items of each historical glue stacking scheme set are reorganized according to the calling weight to obtain several basic process schemes.

[0018] Furthermore, the compatibility of the plurality of dispensing path information and the plurality of lamination geometric information is evaluated respectively, including:

[0019] Extracting interlayer geometric parameters according to the stacking geometric information, and determining geometric process thresholds based on packaging process requirements;

[0020] Calculating the geometric adaptation range of the dispensing path information, obtaining the glue quantity distribution constraint according to the packaging process requirements, and screening the laminate geometric information based on the glue quantity distribution constraint;

[0021] Detecting whether the projection area of ​​the dispensing path information in the stacking packaging direction and the edge spacing of the stacking geometric information meet a preset safety threshold, and obtaining a path conflict detection result based on the spacing detection result;

[0022] According to the geometric adaptation range, the geometric process threshold and the path conflict detection result, a compatibility evaluation result of the dispensing path information and the stacking geometric information is generated.

[0023] Furthermore, the process information of the solution variable items in the basic process solution is adjusted in a targeted manner to obtain several optimized process solutions, including:

[0024] S1: extracting a number of recombinant stacking scheme sets according to the basic process scheme, performing information migration on the recombinant stacking scheme sets based on the compatibility evaluation result, marking schemes that are identical to the basic process scheme after exchange and filtering them out to obtain exchange process schemes, wherein the information migration is the exchange of process information of corresponding scheme variables between different recombinant stacking scheme sets;

[0025] S2: Directively adjusting a single variable item in the exchange process plan, and re-evaluating the compatibility of the exchange process plan after the directive adjustment to obtain an optimized process plan;

[0026] The optimized process plan is used as the optimized basic process plan and steps S1 and S2 are repeated until a preset number of iterations is reached to obtain several optimized process plans.

[0027] Furthermore, setting a call weight according to the compatibility evaluation result and the pre-selected combination scheme includes:

[0028] Based on the compatibility evaluation result, generating a compatibility score for each of the preselected combination solutions;

[0029] Extracting the process success rate of the historical lamination scheme according to the lamination process database, and assigning a basic weight to the variable item of the historical scheme according to the process success rate;

[0030] The compatibility score and the basic weight are linearly weighted to obtain a call weight.

[0031] Furthermore, digital process simulation is performed on each of the preselected stacking schemes to obtain stacking scheme feedback, including:

[0032] Extracting packaging material parameters according to the preselected lamination scheme and constructing a virtual process image, wherein the virtual process image is divided into a geometric structure layer and a process parameter layer;

[0033] Acquire packaging structure parameters according to the geometric structure layer, and acquire glue type and curing conditions according to the process parameter layer;

[0034] Perform glue flow simulation according to the packaging structure parameters, glue type and curing conditions to obtain glue distribution characteristics, and calculate thermomechanical properties based on the process parameter layer;

[0035] The preselected lamination scheme is evaluated according to the glue distribution characteristics and the thermomechanical properties to obtain lamination scheme feedback.

[0036] Further, selecting the corresponding pre-selected lamination scheme as the optimal lamination scheme according to the lamination scheme feedback includes:

[0037] generating a call score based on the glue distribution characteristics and thermomechanical properties;

[0038] Setting a performance evaluation threshold according to historical stacking schemes, comparing the call score with the performance evaluation threshold, and marking the corresponding pre-selected stacking scheme as a stacking candidate scheme if the call score exceeds the performance evaluation threshold;

[0039] The candidate stacking solutions are sorted to obtain the best stacking solution.

[0040] Furthermore, the candidate stacking solutions are sorted to obtain the best stacking solution, including:

[0041] S3: Mark any of the stacking candidate solutions as a sequence base point, compare the remaining stacking candidate solutions with the sequence base point respectively, and adjust the relative positions of the remaining stacking candidate solutions and the sequence base point according to the comparison results to obtain two stacking unordered sets;

[0042] S4: canceling the mark of the stacking candidate solution, and selecting any stacking candidate solution in the stacking unordered set as the sequence base point, and comparing it with the remaining stacking candidate solutions in the stacking unordered set, and adjusting the relative positions of them respectively;

[0043] Repeat steps S3 and S4 until the stacking candidate solutions are sorted and the best stacking solution is obtained.

[0044] An evaluation system for a multi-layer chip stacking solution, the system comprising:

[0045] A variable collection and integration module determines solution variable items and collects data for each of the solution variable items to obtain a plurality of variable information sets, wherein the solution variable items constitute a stacking solution in a stacking packaging process;

[0046] A process plan generating module selects process information corresponding to the variable information set for each of the plan variables and obtains a number of basic process plans;

[0047] A process scheme optimization module adjusts the process information of the scheme variable items in the basic process scheme in a targeted manner to obtain a plurality of optimized process schemes, and uses the plurality of optimized process schemes and the basic process scheme as pre-selected lamination schemes;

[0048] A virtual simulation evaluation module constructs a virtual process image, performs digital process simulation on the pre-selected stacking schemes, and obtains stacking scheme feedback;

[0049] The solution optimization decision module selects the corresponding pre-selected lamination solution as the optimal lamination solution according to the lamination solution feedback.

[0050] Furthermore, the virtual simulation evaluation module includes:

[0051] A virtual image building unit extracts packaging material parameters according to the preselected lamination scheme and builds a virtual process image, wherein the virtual process image is divided into a geometric structure layer and a process parameter layer;

[0052] a parameter extraction and analysis unit, which obtains packaging structure parameters according to the geometric structure layer, and obtains glue type and curing conditions according to the process parameter layer;

[0053] a fluid-thermal coupling simulation unit, which simulates glue flow according to the packaging structure parameters, glue type and curing conditions, obtains glue distribution characteristics, and calculates thermomechanical properties based on the process parameter layer;

[0054] The solution performance evaluation unit evaluates the preselected lamination solution according to the glue distribution characteristics and the thermomechanical properties, and obtains feedback on the lamination solution.

[0055] The technical solution of the present invention can achieve the following technical effects:

[0056] It effectively solves the problems of low debugging efficiency, insufficient utilization of historical data and poor process stability caused by multi-variable coupling in the stacking packaging process. Through data collection, virtual process simulation and iterative optimization, the process parameters of the stacking scheme are dynamically adjusted, which greatly improves the accuracy and adaptability of the process scheme, so that the final selected stacking scheme can more accurately meet different packaging requirements and production environments. This optimization process significantly improves the stability of the packaging process and product quality, reduces the defective rate in the production process, and helps companies reduce production costs and improve production efficiency.

[0057] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention 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 only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 A schematic flow chart of a method for evaluating a multi-layer chip stacking solution;

[0060] Figure 2 Schematic diagram of the process obtained for the basic process plan;

[0061] Figure 3 A flowchart of compatibility assessment is shown below;

[0062] Figure 4 Flow diagram obtained for optimizing process plan;

[0063] Figure 5 Schematic diagram of the structure of the evaluation system for the multi-layer chip stacking solution. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0066] Embodiment 1;

[0067] like Figure 1 As shown, the present application provides a method for evaluating a multi-layer chip stacking solution, the method comprising:

[0068] S100: determining solution variable items and collecting data for each solution variable item to obtain a plurality of variable information sets, wherein the solution variable items constitute a stacking solution in a stacking packaging process;

[0069] S200: For each solution variable item, process information in the corresponding variable information set is selected to obtain several basic process solutions;

[0070] S300: Directly adjusting process information of solution variables in the basic process solution to obtain a number of optimized process solutions, and using the optimized process solutions and the basic process solution as pre-selected stacking solutions;

[0071] S400: Build a virtual process image, perform digital process simulation on pre-selected stacking solutions, and obtain stacking solution feedback;

[0072] S500: Selecting a corresponding pre-selected stacking scheme as an optimal stacking scheme according to stacking scheme feedback.

[0073] Specifically, we first analyze the key factors affecting the performance of the stacked packaging process, such as the characteristics of the adhesive material, the design of the coating path, and other factors, and combine historical process data and actual production needs to determine the solution variables. Then, based on automated equipment, sensors or manual operations, the relevant data of each solution variable item is collected and stored according to a specific format to form several variable information sets; then, each solution variable item selects the corresponding process information according to its data set, and combines these process information to generate several basic process plans. For example, based on the characteristics of the adhesive material, an adhesive material with high temperature resistance or fast curing characteristics can be selected, and based on the coating path information, appropriate coating equipment and coating methods can be selected. On this basis, the variable items of the scheme are adjusted in a targeted manner to generate multiple optimized process schemes. By repeatedly adjusting the process parameters, it is ensured that the final selected stacking scheme can adapt to different packaging requirements; then, a virtual process image is constructed, and the process of the selected pre-selected stacking scheme is digitally simulated. Through simulation, possible problems that may arise in the packaging process can be detected in advance, such as the uniform distribution of glue, interlayer contact and thermal stress, thereby effectively reducing the risk of process failure in production. Finally, based on the feedback results of the virtual process image, the various performance indicators of the stacking scheme, such as glue distribution characteristics and thermomechanical properties, are comprehensively evaluated to select the best stacking scheme.

[0074] The technical solution of the present invention effectively solves the problems of low debugging efficiency, insufficient utilization of historical data and poor process stability caused by multi-variable coupling in the stacking packaging process. Through data acquisition, virtual process simulation and iterative optimization, the process parameters of the stacking scheme are dynamically adjusted, which greatly improves the accuracy and adaptability of the process scheme, so that the final selected stacking scheme can more accurately meet different packaging requirements and production environments. This optimization process significantly improves the stability of the packaging process and product quality, reduces the defective product rate in the production process, and helps enterprises reduce production costs and improve production efficiency.

[0075] Further, if Figure 2 As shown, each solution variable item selects the process information in the corresponding variable information set and obtains several basic process solutions, including:

[0076] S210: Establishing a lamination process database, extracting historical lamination plans based on the lamination process database, and marking plan variables corresponding to the historical lamination plans as pre-selected combination plans, where the plan variables include glue type information, dispensing path information, and lamination geometry information;

[0077] S220: extracting a plurality of dispensing path information and lamination geometry information according to the lamination process database, and performing compatibility evaluation on each dispensing path information and a plurality of lamination geometry information;

[0078] S230: classifying historical lamination schemes according to the glue type information to obtain a plurality of historical lamination scheme sets, wherein each historical lamination scheme set corresponds to each glue type information one-to-one;

[0079] S240: setting a call weight according to the compatibility evaluation result and the pre-selected combination scheme, and reorganizing the process information of the remaining scheme variable items of each historical glue stacking scheme set according to the call weight to obtain several basic process schemes.

[0080] As a preferred embodiment of the above, firstly, various stacking schemes and their related process parameters (such as glue type, dispensing path, stacking geometry information, etc.) used in the historical production process are collected and sorted, the process information of each scheme is classified and marked, and a stacking process database is constructed. The database not only records the correspondence between different glue types and stacking geometry configurations, but also includes data such as the actual performance and success rate of each process scheme. According to the stacking process database, each historical stacking scheme will be marked with its corresponding scheme variable item to form a pre-selected combination scheme. For example, a historical stacking scheme may include the use of a certain type of glue, a specific dispensing path and a specific stacking geometry configuration. Then, several dispensing path information and stacking geometry information are extracted from the stacking process database. The dispensing path information includes the distribution path of the glue during the stacking process, while the stacking geometry information includes the shape, size, arrangement, etc. of the stacking layer; then, each dispensing path information is evaluated for compatibility with multiple stacking geometry information to determine whether they can adapt to each other. For example, a specific dispensing path may require a larger glue coverage area, while certain stacking geometry designs may result in uneven glue coverage, so the two may be incompatible. At this time, the best matching dispensing path and stacking geometry configuration are selected through compatibility evaluation; in order to further optimize the process plan, the historical stacking plans need to be classified according to the glue type. Each type of glue corresponds to a set of historical glue stacking plan sets. These historical glue stacking plan sets contain all stacking plans using this glue type, and each historical plan set contains the dispensing path and stacking geometry information related to this glue type; after obtaining the basic data through the compatibility evaluation results and the pre-selected combination plan, the next step is to set a call weight for each pre-selected combination plan based on the compatibility evaluation results. The weight is used to evaluate the adaptability of each historical stacking plan to ensure that the final selected plan has a higher process success rate and compatibility. Through the call weight, the process information of the remaining plan variable items in each historical glue stacking plan set can be reorganized. Specifically, plans with higher call weights will receive more attention in the next process optimization to ensure that the optimized process plan has better performance. Through the above steps, several basic process plans can be finally obtained.

[0081] Further, if Figure 3 As shown, the compatibility of several dispensing path information and several stacking geometric information is evaluated respectively, including:

[0082] S221: Extracting interlayer geometric parameters based on the stacking geometric information, and determining geometric process thresholds based on packaging process requirements;

[0083] S222: Calculate the geometric adaptation range of the dispensing path information, obtain the glue quantity distribution constraint according to the packaging process requirements, and filter the stacking geometric information based on the glue quantity distribution constraint;

[0084] S223: Detecting whether the projection area of ​​the dispensing path information in the stacking packaging direction and the edge spacing of the stacking geometry information meet a preset safety threshold, and obtaining a path conflict detection result based on the spacing detection result;

[0085] S224: Generate a compatibility evaluation result between the dispensing path information and the stacking geometry information based on the geometric adaptation range, the geometric process threshold, and the path conflict detection result.

[0086] In this embodiment, the geometric parameters of each layer are first extracted from the laminate geometric information. These parameters include the thickness, size, shape and relative position of each layer. Through these geometric parameters, the spacing between each layer and the geometric relationship between layers can be calculated. In some embodiments, the calculation method is as follows: first, the vertical distance between layers (i.e., the interlayer spacing) is obtained by calculating the bottom and top surface positions of two adjacent layers. In addition, the geometric relationship between layers needs to be calculated, such as whether the layers are aligned, whether there is an offset and an angular deviation. The geometric center offset of the layer can be determined by calculating the difference between the geometric center coordinates of the two layers, and the angular relationship is represented by calculating the relative rotation angle. The geometric relationship between layers is obtained based on the calculation results. At the same time, in order to ensure the effectiveness of the stacking process, it is necessary to determine the geometric process threshold based on the specific requirements of the packaging process; for example, in some high-precision packaging processes, in order to avoid glue overflow or uneven distribution due to too small a spacing between layers, it may be necessary to set a minimum interlayer spacing threshold. These thresholds are used to limit the range of geometric parameters to ensure that each layer can be correctly superimposed during the stacking process; then, for each dispensing path, its geometric adaptation range is calculated: a fluid dynamics-based simulation tool can be used to calculate the geometric adaptation range to ensure the optimal fit between the dispensing path and the geometric structure; then, in order to ensure the compatibility of the dispensing path with the stacking geometric information, it is also necessary to detect the projection area of ​​the dispensing path in the stacking packaging direction. If the projection area of ​​the dispensing path is too close to the edge of the stacking geometric information, it may cause glue overflow or uneven coating, thereby affecting the quality of the package; after completing the steps of geometric adaptation range calculation, glue amount distribution constraint screening, path conflict detection, etc., the compatibility evaluation result of the dispensing path information and the stacking geometric information is generated based on all evaluation results.

[0087] Furthermore, if Figure 4 As shown, the process information of the program variables in the basic process plan is adjusted in a targeted manner to obtain several optimized process plans, including:

[0088] S1: Extract several recombinant stacking scheme sets based on the basic process scheme, perform information migration on the recombinant stacking scheme sets based on the compatibility evaluation results, mark the schemes that are identical to the basic process scheme after the exchange and filter them out to obtain the exchange process schemes. Information migration is the exchange of process information of corresponding scheme variables between different recombinant stacking scheme sets;

[0089] S2: Directively adjust the variable item of a single scheme in the exchange process scheme, and re-evaluate the compatibility of the exchange process scheme after the directed adjustment to obtain the optimized process scheme;

[0090] The optimized process plan is used as the optimized basic process plan and steps S1 and S2 are repeated until the preset number of iterations is reached to obtain several optimized process plans.

[0091] Specifically, first, a number of reorganized glue stacking scheme sets are extracted from the basic process scheme, each set includes multiple process variable items related to glue type, dispensing path, stacking geometry, etc. The reorganized glue stacking scheme sets are subjected to information migration based on the results of compatibility evaluation. Information migration refers to the exchange of corresponding process information between different reorganized glue stacking scheme sets. For example, there are two different glue types, and each glue type corresponds to several stacking schemes. According to the compatibility evaluation results, the stacking schemes of different glue types are subjected to migration of dispensing path or stacking geometry information, and there may be schemes that are the same as the basic process scheme. These schemes are then transferred to the reorganized glue stacking scheme set. After screening, several exchange process plans are obtained. After obtaining the exchange process plan, the single scheme variable item in the exchange process plan is adjusted in a targeted manner. The purpose of this step is to make detailed adjustments to the key variable items that may affect the packaging effect in the initially reorganized plan. For example, if the glue flow is uneven or the curing speed is too fast in the matching of the dispensing path and the geometric design, the shape of the dispensing path can be adjusted, or the flow rate of the glue can be adjusted to ensure the stability and accuracy of the packaging process. However, after each adjustment, the compatibility between the variable items of each scheme needs to be re-evaluated to ensure that the adjusted scheme can adapt to the overall process requirements. After the operations of steps S1 and S2, the number of iterations can be set according to historical experience, and steps S1 and S2 can be repeated until the process plan reaches the preset optimization standard or reaches the upper limit of the number of iterations. In each iteration, the optimized process plan will be input into the next round of optimization as a new basic process plan to continuously improve the accuracy and adaptability of the process.

[0092] Furthermore, the call weight is set based on the compatibility evaluation results and the pre-selected combination scheme, including:

[0093] Based on the compatibility assessment results, a compatibility score is generated for each pre-selected combination solution;

[0094] Extract the process success rate of historical stacking plans based on the stacking process database, and assign basic weights to the historical plan variables according to the process success rate;

[0095] The compatibility score and the basic weight are linearly weighted to obtain the call weight.

[0096] As a preferred embodiment of the above, based on the compatibility evaluation results of the variable items of each scheme, each scheme generates a compatibility score according to the matching degree of these parameters. The score reflects the applicability of the scheme and the potential process efficiency. The generation of the compatibility score can utilize advanced analysis algorithms, such as a prediction model based on machine learning; then the process success rate of the historical stacking scheme is extracted from the stacking process database: the extraction of historical success rate can be combined with big data analysis methods. In some embodiments, a statistical analysis method is used to calculate the long-term and short-term success rate of the scheme. For the evaluation of historical processes, a weighted average method can be used to comprehensively consider the influence of various historical data points, and the success rate can be dynamically adjusted according to the changing trend of process parameters and different production conditions. Weighting the success rate under the same production environment, raw materials and equipment conditions can more accurately reflect the adaptability and reliability of each solution. This data shows the performance and reliability of each solution in actual application; each historical solution variable item is assigned a basic weight according to its success rate. This weight reflects the actual effect and credibility of the solution. The determination of the basic weight should take into account the frequency of use of the solution and its performance under different conditions. Statistical analysis methods are used to comprehensively evaluate the long-term and short-term success rates of the solution, as well as its adaptability under different production conditions; finally, the compatibility score is linearly weighted with the basic weight to generate the final call weight. This weight is the key factor in determining whether to adopt a certain solution. Solutions with high weights indicate good compatibility and a high historical success rate.

[0097] Furthermore, digital process simulations are performed on each of the pre-selected stacking solutions to obtain stacking solution feedback, including:

[0098] Extract packaging material parameters based on the pre-selected stacking scheme and build a virtual process image. The virtual process image is divided into a geometric structure layer and a process parameter layer.

[0099] Obtain package structure parameters based on the geometric structure layer, and obtain glue type and curing conditions based on the process parameter layer;

[0100] Perform glue flow simulation based on package structure parameters, glue type, and curing conditions to obtain glue distribution characteristics, and calculate thermomechanical properties based on the process parameter layer;

[0101] Evaluate pre-selected lamination solutions based on glue distribution characteristics and thermo-mechanical performance to obtain lamination solution feedback.

[0102] In this embodiment, according to the pre-selected lamination scheme, the relevant parameters of the packaging material are extracted, mainly including the glue type, curing conditions and the geometric information of the lamination. These parameters are the basis for constructing a virtual process image. The virtual process image includes two parts: the geometric structure layer and the process parameter layer. The geometric structure layer can be generated using a three-dimensional modeling tool, and the process parameter layer can be defined by fluid dynamics simulation and thermodynamics model. The geometric structure layer is used to describe the geometric shape, size, interlayer spacing, arrangement, etc. of the packaging structure. The process parameter layer includes process parameters such as glue type, curing temperature, and curing time. These parameters directly affect the fluidity and curing process of the glue, and thus affect the overall performance of the package. The detailed parameters of the packaging structure are then extracted through the geometric structure layer, such as the thickness, size, alignment, interlayer spacing and other information of each layer. At the same time, the glue type (such as epoxy resin, polyurethane, etc.) and its curing conditions (curing temperature, curing time, etc.) are obtained from the process parameter layer. Based on the packaging structure parameters and process parameters The glue flow simulation and thermomechanical properties calculation are carried out based on the information of several layers: the glue flow simulation uses computational fluid dynamics technology to simulate the flow of glue: a three-dimensional geometric model is established and the physical properties and boundary conditions of the glue are defined, and then meshing is performed and the flow equation is solved to analyze the flow rate, pressure distribution and other data of the glue. By checking the uniformity of the glue distribution in each area, the problem of uneven flow is found, and the dispensing path, injection parameters or geometric design are adjusted according to the results to optimize the uniform distribution and fluidity of the glue, thereby ensuring the effectiveness of the final packaging process; the thermomechanical properties calculation evaluates the temperature distribution, thermal expansion and stress changes during the glue curing process through thermodynamic simulation and mechanical stress analysis: thermal analysis is performed based on the curing conditions to calculate the temperature field during the glue curing process to ensure uniform temperature distribution. Then, thermal expansion analysis is used to evaluate the thermal expansion difference between the glue and the packaging material, calculate the stress distribution generated during the curing process, and avoid package deformation or cracks caused by thermal stress. Finally, finite element analysis is used to comprehensively consider thermal stress and deformation, optimize process parameters, and ensure the stability and reliability of the package structure. Then, based on the results of glue flow simulation and the calculation of thermomechanical properties, the overall performance of the preselected stacking solution is evaluated to assess whether the glue can be evenly distributed between the layers of the stacking structure, avoiding package failure caused by uneven glue distribution. The uniformity of glue distribution directly affects the bonding effect and overall strength of the package. The mechanical stress generated by thermal expansion and temperature changes during the curing process is evaluated to ensure that the stacking structure remains stable after curing and does not fail due to excessive thermal stress. Based on these evaluation results, feedback on the stacking solution is generated.

[0103] Furthermore, according to the stacking scheme feedback, a corresponding pre-selected stacking scheme is selected as the optimal stacking scheme, including:

[0104] Generate call scores based on glue distribution characteristics and thermomechanical properties;

[0105] A performance evaluation threshold is set based on historical stacking schemes. The call score is compared with the performance evaluation threshold. If the call score exceeds the performance evaluation threshold, the corresponding pre-selected stacking scheme is marked as a stacking candidate scheme.

[0106] Sort several stacking candidate solutions to obtain the best stacking solution.

[0107] Specifically, based on the comprehensive evaluation of glue distribution characteristics and thermomechanical properties, a glue distribution score and a thermomechanical performance score are generated for each pre-selected stacking scheme. The call score can be generated by using a weighted average method to perform a weighted average of the glue distribution characteristics and the thermomechanical performance scores. The weights of the two are set according to the importance of the process requirements. For example, if for a certain packaging process, glue uniformity is more important than thermomechanical performance, a higher weight is given to the glue distribution characteristics. The calculation formula for the call score is:

[0108] ;

[0109] in, and are the weight coefficients of glue distribution characteristics and thermomechanical properties respectively; then, according to the performance data of historical stacking schemes, a performance evaluation threshold is set, which is set by analyzing historical successful schemes and extracting their performance data, and comparing the call score of each pre-selected stacking scheme with the set performance evaluation threshold. If the call score exceeds the performance evaluation threshold, the pre-selected stacking scheme is marked as a "stacking candidate scheme", which means that the scheme performs well in terms of glue distribution characteristics and thermomechanical properties, has strong adaptability, and meets the process requirements; once multiple stacking candidate schemes are determined, the multiple stacking candidate schemes are sorted, and the best stacking scheme is obtained according to the sorting results.

[0110] Furthermore, the stacking candidate solutions are sorted to obtain the best stacking solution, including:

[0111] S3: Mark any stacking candidate solution as a sequence base point, compare the remaining stacking candidate solutions with the sequence base point respectively, and adjust the relative positions of the remaining stacking candidate solutions and the sequence base point according to the comparison results to obtain two stacking unordered sets;

[0112] S4: cancel the mark of the stacking candidate solution, select any stacking candidate solution in the stacking unordered set as the sequence base point, compare it with the remaining stacking candidate solutions in the stacking unordered set, and adjust the relative positions respectively;

[0113] Repeat steps S3 and S4 until the stacking candidate solutions are sorted and the best stacking solution is obtained.

[0114] As a preferred embodiment of the above, one candidate stacking solution is selected from all candidate stacking solutions as the sequence base point, and the remaining candidate stacking solutions are compared one by one with this base point. By comparing the correlation and adaptability of each candidate solution with the base point, their relative positions are adjusted to form two stacking unordered sets. In this case, all candidate solutions not selected as the base point are considered elements of the unordered set. The comparison criteria can be glue distribution, thermomechanical properties, or other evaluation criteria. Using these criteria, the difference between each candidate solution and the sequence base point is determined, and its relative position is adjusted based on the degree of difference. Through comparison and adjustment, each candidate solution is optimized relative to the sequence base point, ultimately forming two separate unordered sets: one containing solutions that are more similar to the base point, and the other containing solutions that differ significantly from the base point. When adjusting the relative positions, the multi-dimensional evaluation results of each solution (such as compatibility score, process performance, production cost, etc.) are considered. Multi-attribute decision-making methods (such as the analytic hierarchy process (AHP)) can be combined to more scientifically adjust the ranking of the candidate solutions. The current sequence base point is then unmarked, and a new solution from the stacking unordered set is selected as the new sequence base point. The new base point is then compared with the remaining candidate solutions, and their relative positions are adjusted to form a new unordered set. In this way, the comparison and adjustment steps are repeated until all candidate solutions are sorted according to their relative adaptability. When updating the base point, the new sequence base point is randomly called from the unordered set. With each adjustment, the relative position of the candidate solution is gradually optimized. Ultimately, all candidate solutions are sorted according to the evaluation criteria. After multiple rounds of iterative adjustments, all stacking candidate solutions will eventually be sorted, resulting in a sorted result. The final optimal stacking solution is obtained based on the sorted result.

[0115] Embodiment 2;

[0116] Based on the same inventive concept as the evaluation method for a multi-layer chip stacking solution in the aforementioned embodiment, the present invention also provides an evaluation system for a multi-layer chip stacking solution, such as Figure 5 As shown, the system includes:

[0117] The variable collection and integration module determines the program variable items and collects data for each program variable item to obtain several variable information sets. The program variable items constitute the stacking program in the stacking packaging process;

[0118] In the process plan generation module, each plan variable item selects the process information in the corresponding variable information set and obtains several basic process plans;

[0119] The process plan optimization module adjusts the process information of the plan variables in the basic process plan in a targeted manner to obtain several optimized process plans, and uses the several optimized process plans and the basic process plan as pre-selected stacking plans;

[0120] The virtual simulation evaluation module builds a virtual process image, performs digital process simulation on pre-selected stacking solutions, and obtains feedback on the stacking solutions.

[0121] The solution optimization decision module selects the corresponding pre-selected stacking solution as the optimal stacking solution based on the stacking solution feedback.

[0122] The above-mentioned adjustment system in the present invention can effectively implement the evaluation method of the multi-layer chip stacking solution, and the technical effects that can be achieved are as described in the above-mentioned embodiments and will not be repeated here.

[0123] More specifically, the virtual simulation assessment module includes:

[0124] The virtual image building unit extracts packaging material parameters based on the pre-selected stacking scheme and builds a virtual process image. The virtual process image is divided into a geometric structure layer and a process parameter layer.

[0125] Parameter extraction and analysis unit, which obtains package structure parameters based on the geometric structure layer, and obtains glue type and curing conditions based on the process parameter layer;

[0126] The fluid-thermal coupling simulation unit simulates the glue flow according to the packaging structure parameters, glue type and curing conditions, obtains the glue distribution characteristics, and calculates the thermomechanical properties based on the process parameter layer;

[0127] The solution performance evaluation unit evaluates the pre-selected stacking solution based on the glue distribution characteristics and thermo-mechanical properties, and obtains feedback on the stacking solution.

[0128] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the corresponding optimization effects of the method in Example 1, which will not be repeated here.

[0129] Although the present application has been described with reference to specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and drawings are merely illustrative of the present application as defined herein and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, the present application is intended to include such modifications and variations as fall within the scope of the present application and its equivalents.

Claims

1. A method for evaluating a multi-layer chip stacking solution, characterized in that: The method comprises: Determine solution variable items, and collect data for each solution variable item to obtain a plurality of variable information sets, wherein the solution variable items constitute a stacking solution in a stacking packaging process; Each of the scheme variable items selects the process information corresponding to the variable information set and obtains several basic process schemes, including: Establishing a lamination process database, extracting historical lamination schemes according to the lamination process database, and marking the scheme variable items corresponding to the historical lamination schemes as pre-selected combination schemes, wherein the scheme variable items include glue type information, dispensing path information, and lamination geometry information; Extracting a plurality of dispensing path information and lamination geometry information according to the lamination process database, and performing compatibility evaluation on each dispensing path information and a plurality of lamination geometry information; Classifying the historical lamination schemes according to the glue type information to obtain a plurality of historical lamination scheme sets, wherein each of the historical lamination scheme sets corresponds to each of the glue type information; Setting a call weight according to the compatibility evaluation result and the pre-selected combination scheme, and reorganizing the process information of the remaining scheme variables of each of the historical glue-stack scheme sets according to the call weight to obtain a number of basic process schemes; Directively adjusting the process information of the solution variable items in the basic process solution to obtain a plurality of optimized process solutions, and using the plurality of optimized process solutions and the basic process solution as pre-selected lamination solutions; Construct a virtual process mirror, perform digital process simulation on the pre-selected stacking schemes, and obtain stacking scheme feedback; The corresponding preselected lamination scheme is selected as the optimal lamination scheme according to the lamination scheme feedback.

2. The method for evaluating a multi-layer chip stacking solution according to claim 1, wherein: Compatibility evaluation is performed on the plurality of dispensing path information and the plurality of lamination geometry information, including: Extracting interlayer geometric parameters according to the stacking geometric information, and determining geometric process thresholds based on packaging process requirements; Calculating the geometric adaptation range of the dispensing path information, obtaining the glue quantity distribution constraint according to the packaging process requirements, and screening the laminate geometric information based on the glue quantity distribution constraint; Detecting whether the projection area of ​​the dispensing path information in the stacking packaging direction and the edge spacing of the stacking geometric information meet a preset safety threshold, and obtaining a path conflict detection result based on the spacing detection result; According to the geometric adaptation range, the geometric process threshold and the path conflict detection result, a compatibility evaluation result of the dispensing path information and the stacking geometric information is generated.

3. The method for evaluating a multi-layer chip stacking solution according to claim 1, wherein: Directedly adjust the process information of the solution variable items in the basic process solution to obtain several optimized process solutions, including: S1: extracting a number of recombinant stacking scheme sets according to the basic process scheme, performing information migration on the recombinant stacking scheme sets based on the compatibility evaluation result, marking schemes that are identical to the basic process scheme after exchange and filtering them out to obtain exchange process schemes, wherein the information migration is the exchange of process information of corresponding scheme variables between different recombinant stacking scheme sets; S2: Directively adjusting a single variable item in the exchange process plan, and re-evaluating the compatibility of the exchange process plan after the directive adjustment to obtain an optimized process plan; The optimized process plan is used as the optimized basic process plan and steps S1 and S2 are repeated until a preset number of iterations is reached to obtain several optimized process plans.

4. The method for evaluating a multi-layer chip stacking solution according to claim 1, wherein: The call weight is set according to the compatibility evaluation result and the pre-selected combination scheme, including: Based on the compatibility evaluation result, generating a compatibility score for each of the preselected combination solutions; Extracting the process success rate of the historical lamination scheme according to the lamination process database, and assigning a basic weight to the variable item of the historical scheme according to the process success rate; The compatibility score and the basic weight are linearly weighted to obtain a call weight.

5. The method for evaluating a multi-layer chip stacking solution according to claim 1, wherein: Performing process digital simulation on each of the preselected stacking schemes to obtain stacking scheme feedback, including: Extracting packaging material parameters according to the preselected lamination scheme and constructing a virtual process image, wherein the virtual process image is divided into a geometric structure layer and a process parameter layer; Acquire packaging structure parameters according to the geometric structure layer, and acquire glue type and curing conditions according to the process parameter layer; Perform glue flow simulation according to the packaging structure parameters, glue type and curing conditions to obtain glue distribution characteristics, and calculate thermomechanical properties based on the process parameter layer; The preselected lamination scheme is evaluated according to the glue distribution characteristics and the thermomechanical properties to obtain lamination scheme feedback.

6. The method for evaluating a multi-layer chip stacking solution according to claim 5, wherein: Selecting the corresponding pre-selected lamination scheme as the optimal lamination scheme according to the lamination scheme feedback includes: generating a call score based on the glue distribution characteristics and thermomechanical properties; Setting a performance evaluation threshold according to historical stacking schemes, comparing the call score with the performance evaluation threshold, and marking the corresponding pre-selected stacking scheme as a stacking candidate scheme if the call score exceeds the performance evaluation threshold; The candidate stacking solutions are sorted to obtain the best stacking solution.

7. The method for evaluating a multi-layer chip stacking solution according to claim 6, wherein: Sorting the candidate stacking solutions to obtain the best stacking solution includes: S3: Mark any of the stacking candidate solutions as a sequence base point, compare the remaining stacking candidate solutions with the sequence base point respectively, and adjust the relative positions of the remaining stacking candidate solutions and the sequence base point according to the comparison results to obtain two stacking unordered sets; S4: canceling the mark of the stacking candidate solution, and selecting any stacking candidate solution in the stacking unordered set as the sequence base point, and comparing it with the remaining stacking candidate solutions in the stacking unordered set, and adjusting the relative positions of them respectively; Repeat steps S3 and S4 until the stacking candidate solutions are sorted and the best stacking solution is obtained.

8. An evaluation system for a multi-layer chip stacking solution, characterized in that: The method for evaluating a multi-layer chip stacking solution according to claim 1 , wherein the system comprises: A variable collection and integration module determines solution variable items and collects data for each of the solution variable items to obtain a plurality of variable information sets, wherein the solution variable items constitute a stacking solution in a stacking packaging process; A process plan generating module selects process information corresponding to the variable information set for each of the plan variables and obtains a number of basic process plans; A process scheme optimization module adjusts the process information of the scheme variable items in the basic process scheme in a targeted manner to obtain a plurality of optimized process schemes, and uses the plurality of optimized process schemes and the basic process scheme as pre-selected lamination schemes; A virtual simulation evaluation module constructs a virtual process image, performs digital process simulation on the pre-selected stacking schemes, and obtains stacking scheme feedback; The solution optimization decision module selects the corresponding pre-selected lamination solution as the optimal lamination solution according to the lamination solution feedback.

9. The multi-layer chip stacking solution evaluation system according to claim 8, characterized in that: The virtual simulation evaluation module includes: A virtual image building unit extracts packaging material parameters according to the preselected lamination scheme and builds a virtual process image, wherein the virtual process image is divided into a geometric structure layer and a process parameter layer; a parameter extraction and analysis unit, which obtains packaging structure parameters according to the geometric structure layer, and obtains glue type and curing conditions according to the process parameter layer; a fluid-thermal coupling simulation unit, which simulates glue flow according to the packaging structure parameters, glue type and curing conditions, obtains glue distribution characteristics, and calculates thermomechanical properties based on the process parameter layer; The solution performance evaluation unit evaluates the preselected lamination solution according to the glue distribution characteristics and the thermomechanical properties, and obtains feedback on the lamination solution.

Images