An FC substrate plastic package warpage optimization method

Abstract

The present application relates to chip plastic sealing technical field, especially to a kind of FC substrate plastic sealing warping optimization method, comprising: obtaining FC substrate initial lamination, packaging structure and epoxy plastic sealing material thermal mechanical and solidification shrinkage parameters, establish packaging structure simulation model;Under the same process condition, respectively carry out only the first simulation considering thermal strain and the second simulation superimposed solidification shrinkage, obtain solidification shrinkage out-of-plane displacement component and initial predicted warping value by comparison;The first out-of-plane displacement distribution is combined with the component to determine the plastic sealing bending moment direction, lock the material layer to be adjusted on one side of the center surface of substrate thickness, form compensation bending moment opposite to plastic sealing bending moment by adjusting its material type or layer thickness, generate candidate lamination design;Packaging deformation simulation is carried out on candidate scheme, and scheme with predicted warping value less than initial value is selected as target lamination. The influence of thermal deformation and solidification shrinkage can be separated by the present application, and the adjustment direction misjudgment caused by the superposition of the two is avoided.

Description

Technical Field

[0001] This invention relates to the technical field of chip molding, and more particularly to a method for optimizing warpage during molding of FC substrates. Background Technology

[0002] With the development of semiconductor packaging technology, flip-chip packaging structures are gradually evolving towards higher integration, multilayering, and thinner designs. FC substrates typically include multiple conductor layers, dielectric layers, and solder mask layers, forming the packaging structure together with the chip and epoxy molding compound. During the packaging process, the epoxy molding compound undergoes heating, curing, and cooling. The differences in thermal expansion between the molding compound and the substrate, chip, and other materials, as well as the volume changes during curing, affect the stress distribution and deformation state of the packaging structure.

[0003] Existing technologies typically assess the warpage of packaged structures after molding through experimental testing or finite element simulation. Finite element simulation can predict the overall deformation of the packaged structure based on its geometric parameters, material parameters, and process conditions, and provide a reference for the stack-up design of FC substrates.

[0004] However, the overall deformation of the packaging structure is the result of the combined effect of the thermomechanical response of each component material and the curing shrinkage of the epoxy molding compound. Different deformation factors may have different directions and degrees of influence on the external deformation. When adjusting the FC substrate stack based on the overall deformation results, it is difficult to determine the adjustment direction of the FC substrate stack in a targeted manner. Summary of the Invention

[0005] This invention provides a method for optimizing the warpage of FC substrate molding, which can effectively solve the problems in the background art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for optimizing warpage during molding of an FC substrate includes: Acquire the initial stack-up data of the FC substrate, the encapsulation structure data, and the thermomechanical parameters and curing shrinkage parameters of the epoxy molding compound, and establish a simulation model of the encapsulation structure including the FC substrate, chip, and epoxy molding compound. Under the same molding process conditions, the packaging structure simulation model is subjected to a first simulation that only considers thermal strain and a second simulation that simultaneously considers thermal strain and curing shrinkage strain to obtain the first and second out-of-plane displacement distributions. The curing shrinkage out-of-plane displacement component is determined based on the difference between the two, and the initial predicted warpage value is determined based on the second out-of-plane displacement distribution. The direction of the molding bending moment is determined based on the first out-of-plane displacement distribution and the out-of-plane displacement component of the curing shrinkage. Based on this, the material layer to be adjusted on one side of the thickness center plane of the FC substrate is determined. At least one of the material type and layer thickness of the material layer to be adjusted is adjusted so that the adjusted stack forms a compensating bending moment opposite to the direction of the molding bending moment, thus obtaining the candidate stack design. Under the molding process conditions, the candidate stack design was simulated for encapsulation deformation, and the candidate stack design with a predicted warpage value less than the initial predicted warpage value was determined as the target stack design.

[0007] Furthermore, the initial stack-up data includes the stack-up order, material type, layer thickness, and conductor pattern distribution of each conductor layer, dielectric layer, and solder mask layer in the FC substrate. The packaging structure data includes the geometric dimensions and relative positions of the FC substrate, chip, and epoxy encapsulation.

[0008] Furthermore, before performing the first and second simulations, the measured warpage value of the encapsulation structure formed using a known stack design and a known epoxy molding compound under known molding process conditions is obtained, and the encapsulation structure simulation model is used to perform simulations including thermal strain and curing shrinkage strain to obtain the corresponding simulated warpage value. Based on the absolute value of the difference between the measured warpage value and the simulated warpage value, adjust at least one of the material parameters and interface constraint parameters of the packaging structure simulation model until the absolute value is not greater than the preset correction threshold.

[0009] Furthermore, the thermomechanical parameters and curing shrinkage parameters of multiple candidate epoxy molding compounds were obtained; While keeping the target stack design and molding process conditions unchanged, the encapsulation deformation simulation was performed using the thermomechanical parameters and curing shrinkage parameters corresponding to each candidate epoxy molding compound. The candidate epoxy molding compound with the smallest predicted warpage value was then determined as the target epoxy molding compound that matches the target stack design.

[0010] Furthermore, the target stack design determined in the current design iteration is used as the initial stack data for the next design iteration. The determination of the material layer to be adjusted, the formation of candidate stack designs, and the encapsulation deformation simulation are repeated until the predicted warpage value corresponding to the target stack design is not greater than the preset warpage threshold.

[0011] Furthermore, the first and second simulations used the same packaging structure, boundary constraints, and temperature history; In the first simulation, no curing shrinkage strain is applied. In the second simulation, curing shrinkage strain is applied, and the out-of-plane displacement values ​​of the same model coordinate positions in the first out-of-plane displacement distribution are subtracted from the out-of-plane displacement values ​​of each model coordinate position in the second out-of-plane displacement distribution to obtain the curing shrinkage out-of-plane displacement components.

[0012] Furthermore, a first direction and a second direction that are perpendicular to each other are determined in a plane parallel to the FC substrate; Based on the out-of-plane displacement distribution, the thermal strain curvature components in the first and second directions are determined respectively. Based on the out-of-plane displacement components of curing shrinkage, the curvature components of curing shrinkage in the first and second directions are determined respectively. The curvature components of thermal strain and curvature shrinkage in the same direction are added together to obtain the total curvature component in the corresponding direction. The direction of the molding bending moment in the corresponding direction is determined according to the positive and negative directions of the total curvature component.

[0013] Furthermore, each material layer located on both sides of the thickness center plane of the FC substrate and not crossing the thickness center plane is defined as a test material layer; For each test material layer, a test stack design is formed that adjusts at least one of the material type and layer thickness of the test material layer, and the test stack design is encapsulated and deformed. Based on the predicted curvature direction corresponding to the trial layer design, the trial material layer whose predicted curvature direction is opposite to the direction of the molding bending moment is determined as the material layer to be adjusted.

[0014] Furthermore, while keeping the thickness of the material layer to be adjusted constant, the material layer to be adjusted is replaced with a candidate material. At least one of the elastic modulus and thermal expansion coefficient of the candidate material is different from the corresponding parameter of the material layer to be adjusted in the initial stack data, so that the candidate stack design using the candidate material forms a compensating bending moment.

[0015] Furthermore, while keeping the material type of the material layer to be adjusted unchanged, the thickness of the material layer to be adjusted is changed, and another material layer in the FC substrate is adjusted by a thickness change that is equal in magnitude and opposite in direction to the thickness change of the material layer to be adjusted, so that the total thickness of the substrate of the candidate stack design is the same as the total thickness of the substrate corresponding to the initial stack data.

[0016] The technical solution of this invention can achieve the following technical effects: by decoupling the independent contributions of thermal strain and curing shrinkage strain to package warpage, the dominant direction of the molding bending moment can be identified, avoiding misjudgment of the adjustment direction under multi-factor coupling; based on the bending moment balance principle, the properties or thickness of specific material layers can be directionally adjusted to construct a reverse compensation bending moment inside the stack, suppressing overall deformation from a mechanical mechanism perspective; combined with simulation correction and iterative optimization mechanisms, the reliability of FC substrate stack design can be improved, and the molding warpage amplitude can be reduced.

[0017] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the FC substrate molding warpage optimization method in an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0022] like Figure 1 As shown, the FC substrate molding warpage optimization method of the present invention specifically includes the following steps: Step S100: Obtain the initial stacking data of the FC substrate, the packaging structure data, and the thermomechanical parameters and curing shrinkage parameters of the epoxy molding compound, and establish a packaging structure simulation model including the FC substrate, chip and epoxy molding compound; Step S200: Under the same molding process conditions, perform a first simulation that only considers thermal strain and a second simulation that simultaneously considers thermal strain and curing shrinkage strain on the packaging structure simulation model to obtain the first out-of-plane displacement distribution and the second out-of-plane displacement distribution. Determine the curing shrinkage out-of-plane displacement component based on the difference between the first out-of-plane displacement distribution and the second out-of-plane displacement distribution, and determine the initial predicted warpage value based on the second out-of-plane displacement distribution. Step S300: Determine the direction of the molding bending moment based on the first out-of-plane displacement distribution and the out-of-plane displacement component of the curing shrinkage, and determine the material layer to be adjusted on one side of the thickness center plane of the FC substrate accordingly. Adjust at least one of the material type and layer thickness of the material layer to be adjusted so that the adjusted stack forms a compensating bending moment opposite to the direction of the molding bending moment, and obtain the candidate stack design. Step S400: Perform encapsulation deformation simulation on the candidate stack-up design under the molding process conditions, and determine the candidate stack-up design with a predicted warpage value less than the initial predicted warpage value as the target stack-up design.

[0023] In this embodiment, the out-of-plane displacement components caused by thermal strain and curing shrinkage strain are separated through the first and second simulations, respectively obtaining the thermal strain curvature component and the curing shrinkage curvature component. This allows the contributions of the two deformation factors with different physical mechanisms to be independently quantified and used to determine the stack adjustment direction. The molding bending moment direction is determined based on the total curvature direction obtained by superimposing the thermal strain curvature component and the curing shrinkage curvature component, which avoids misjudgment of the adjustment direction due to the opposite or partial cancellation of the effects of thermal deformation and curing shrinkage deformation. By selecting a material layer located on one side of the thickness center plane of the FC substrate and not crossing the thickness center plane for adjustment, the material type or thickness change of this single-sided material layer forms a compensating bending moment opposite to the molding bending moment direction. This achieves warpage suppression while keeping the number of substrate conductor layers, the position of the connection pads, and the electrical connection relationship unchanged, avoiding the impact of stack adjustment on the electrical performance of the substrate. The above steps form a closed-loop feedback relationship between simulation, analysis, and adjustment. Without changing the packaging process conditions and chip design, out-of-plane deformation after molding can be suppressed simply by adjusting the stack parameters inside the FC substrate.

[0024] In a specific implementation, as one example, step S100 is used to establish a packaging structure simulation model to obtain a computational model that can reflect the actual thermo-mechanical coupling behavior of the FC substrate, chip, and epoxy encapsulation. If theoretical material parameters are directly used for simulation, due to the deviation between the actual material properties and nominal values, and the difficulty in accurately quantifying the interface contact state, the simulation results often deviate significantly from the measured warpage values. Therefore, the layer adjustments based on the simulation results will lose a reliable quantitative benchmark. This embodiment corrects the simulation model by introducing measured warpage values, making the output of the simulation model under the combined action of thermal strain and curing shrinkage strain more consistent with the measured results. The specific implementation is as follows: Step S110: Obtain the initial stack-up data, encapsulation structure data, and thermomechanical parameters and curing shrinkage parameters of the epoxy molding compound for the FC substrate; The initial stack-up data of the FC substrate is extracted from the design file of the FC substrate. This design file is the stack-up design file or Gerber file used in the manufacturing stage of the FC substrate. It records the stack-up order of each conductor layer, dielectric layer and solder mask layer in the FC substrate, the material type of each layer, the thickness of each layer and the conductor pattern distribution of each conductor layer. The conductor pattern distribution refers to the planar layout of conductive structures such as lines, pads and vias in each conductor layer. The packaging structure data is extracted from the packaging design documents, which include dimensional drawings of the FC substrate, dimensional drawings and mounting drawings of the chip, and molding area definition documents of the epoxy molding compound. These documents record the geometric dimensions of the FC substrate, the chip, the epoxy molding compound, and the relative positional relationships between the FC substrate, the chip, and the epoxy molding compound. The relative positional relationships include the position coordinates of the chip flip-chip connection to the upper surface of the FC substrate and the area of ​​the epoxy molding compound that encapsulates the chip and covers the upper surface of the FC substrate. The thermomechanical parameters of epoxy molding compounds are obtained from the material data sheets provided by the epoxy molding compound supplier. These material data sheets include the elastic modulus as a function of temperature, the glass transition temperature, the first coefficient of thermal expansion at a temperature not exceeding the glass transition temperature, and the second coefficient of thermal expansion at a temperature exceeding the glass transition temperature. Epoxy molding compounds exhibit different thermal expansion behaviors before and after the glass transition temperature. The first coefficient of thermal expansion corresponds to the thermal expansion rate in the glassy state, and the second coefficient of thermal expansion corresponds to the thermal expansion rate in the rubbery state. The curing shrinkage parameters of epoxy molding compounds are obtained from the curing shrinkage rate data provided by the supplier. These curing shrinkage parameters are used to characterize the volume shrinkage or linear shrinkage that occurs during the curing process of the epoxy molding compound. They are usually provided in the form of linear shrinkage rate, which is related to the chemical reaction shrinkage and volume change during the cooling process of the epoxy molding compound. Step S120: Based on the acquired initial stack-up data of the FC substrate, packaging structure data, and thermomechanical parameters and curing shrinkage parameters of the epoxy molding compound, a packaging structure simulation model including the FC substrate, chip, and epoxy molding compound is established. This packaging structure simulation model is established in a finite element simulation environment. Each conductor layer, dielectric layer, and solder mask layer of the FC substrate is constructed according to the stacking order and layer thickness in the initial stack-up data, and each layer material is assigned corresponding material properties according to its material type. The chip and epoxy molding compound are constructed according to the geometric dimensions and relative positions in the packaging structure data. The connection structure between the chip and the FC substrate is treated as corresponding contact or connection constraints in the simulation model. When constructing the packaging structure simulation model, boundary constraints are applied to the model to simulate the support conditions of the packaging structure during the molding process. The boundary constraints restrict the rigid body displacement of the model in three-dimensional space, while allowing the structure to deform freely under temperature changes and curing shrinkage. The mesh division of the model is determined according to the geometric characteristics and deformation gradient of each component structure, and a finer mesh is used in areas with high conductor pattern distribution density and near the interface. Step S130: Based on the established packaging structure simulation model, obtain the measured warpage value of the packaging structure formed using a known stack-up design and a known epoxy molding compound under known molding process conditions. The measured warpage value is obtained through a non-contact optical measurement method, such as using the shadow moiré method or digital image correlation method to measure the out-of-plane displacement of the packaged structure. During the measurement, a preset reference area on the lower surface of the packaging structure is used as a reference to obtain the overall warpage shape and its maximum warpage value of the packaging structure. The preset reference area is the area on the lower surface of the FC substrate where the out-of-plane displacement is constrained or the deformation is minimal during the packaging process. Typically, the central area of ​​the lower surface of the FC substrate or the area near the support position is selected. Step S140: Using the established encapsulation structure simulation model, a simulation including thermal strain and curing shrinkage strain is performed under known molding process conditions to obtain the corresponding simulated warpage value. In this simulation, the temperature history is applied according to the temperature curve of the known molding process conditions. The elastic modulus of the epoxy encapsulation changes with temperature, and a corresponding elastic modulus value is assigned according to the simulation temperature at each step. When the temperature of the epoxy encapsulation is not higher than its glass transition temperature, the first thermal expansion coefficient is used. When the temperature of the epoxy encapsulation is higher than its glass transition temperature, the second thermal expansion coefficient is used. The curing shrinkage strain is applied in the simulation according to the curing shrinkage parameters of the epoxy encapsulation material as an initial strain. After the simulation is completed, the out-of-plane displacement distribution of the lower surface of the encapsulation structure is extracted from the simulation results. A warpage reference plane is fitted based on a preset reference region. The difference between the maximum and minimum out-of-plane displacement values ​​relative to the warpage reference plane in the out-of-plane displacement distribution is determined as the simulated warpage value. Step S150: Compare the obtained measured warpage value with the simulated warpage value, calculate the absolute value of the difference between the two, and compare it with a preset correction threshold. The value of this threshold is determined based on the measurement accuracy of the measured warpage value and the engineering accuracy requirements of the simulation model. When the absolute value is greater than the preset correction threshold, adjust at least one of the material parameters and interface constraint parameters of the packaging structure simulation model, return to step S140 to re-perform the simulation including thermal strain and curing shrinkage strain, obtain the updated simulated warpage value, and recalculate the absolute value of the difference between the updated simulated warpage value and the measured warpage value. Repeat the above process until the absolute value is not greater than the preset correction threshold. Among them, the adjustment of material parameters includes, but is not limited to, scaling of elastic modulus, correction of thermal expansion coefficient, and adjustment of curing shrinkage rate. The adjustment of interface constraint parameters includes, but is not limited to, the contact stiffness coefficient between each layer of the FC substrate, the connection stiffness coefficient between the chip and the FC substrate, and the interface friction coefficient or adhesive strength coefficient between the epoxy encapsulator and the FC substrate and the chip. When the absolute value is not greater than the preset correction threshold, the currently adjusted packaging structure simulation model is determined as the corrected simulation model.

[0025] In this embodiment, the initial stack-up data of the FC substrate is used to construct the geometric model of each layer of the FC substrate and assign corresponding material properties; the packaging structure data is used to determine the geometric dimensions and relative assembly positions of the FC substrate, chip, and epoxy molding compound in the simulation space; the thermomechanical parameters of the epoxy molding compound are used to assign the elastic modulus and coefficient of thermal expansion of the epoxy molding compound as a function of temperature in each incremental step of the simulation temperature history; the curing shrinkage parameters of the epoxy molding compound are used to apply the volume shrinkage effect during the curing process to the epoxy molding compound in the form of initial strain; the above data together constitute the complete input of the packaging structure simulation model; on this basis, the packaging structure simulation model is corrected by the measured warpage value, so that the deviation between the output under the combined action of thermal strain and curing shrinkage strain and the measured result is controlled within the preset correction threshold range, and the simulation model tends to be consistent with the measured result in terms of overall deformation amplitude and distribution pattern; the simulation based on this correction model can provide a deformation distribution consistent with the real packaging structure, reducing the interference of simulation errors caused by material parameter deviations and interface state uncertainties on the stack-up adjustment direction.

[0026] In a specific implementation, as one example, step S200 is used to separate the out-of-plane displacement component caused solely by curing shrinkage strain from the deformation output of the encapsulation structure simulation model under temperature history, and to determine the initial predicted warpage value. Since both thermal strain and curing shrinkage strain act on the epoxy encapsulation body in the form of strain fields in the simulation model, if only one simulation containing both strains is performed, their contributions will be superimposed in the same displacement field and cannot be distinguished. This embodiment sets up two control cases: a first simulation considering only thermal strain and a second simulation considering both thermal strain and curing shrinkage strain. The out-of-plane displacement distribution of the second simulation is subtracted from the out-of-plane displacement distribution of the first simulation, and the difference is the curing shrinkage out-of-plane displacement component. At the same time, the initial predicted warpage value is determined based on the out-of-plane displacement distribution of the second simulation. The specific implementation is as follows: Step S210: Under the same molding process conditions, perform a first simulation on the corrected packaging structure simulation model, considering only thermal strain. The temperature history applied in the first simulation is consistent with the temperature curve in the actual molding process conditions. In each incremental step of the simulation temperature history, assign a corresponding elastic modulus value to the epoxy molding compound based on the temperature of the current incremental step, and select the coefficient of thermal expansion according to the relationship between the temperature of the epoxy molding compound and its glass transition temperature: when the temperature of the epoxy molding compound is not higher than its glass transition temperature, the first coefficient of thermal expansion is used; when the temperature of the epoxy molding compound is higher than its glass transition temperature, the second coefficient of thermal expansion is used. The coefficient of thermal expansion and the elastic modulus that change with temperature of each layer of the FC substrate are also assigned corresponding values ​​in each incremental step. No curing shrinkage strain is applied in the first simulation. The packaging structure generates thermal strain due to the difference in the coefficients of thermal expansion of each component material under the action of temperature change. This thermal strain causes the packaging structure to undergo out-of-plane deformation. After the first simulation is completed, extract the out-of-plane displacement values ​​at the coordinate positions of each model on the lower surface of the packaging structure to form the first out-of-plane displacement distribution. Step S220: Under the same molding process conditions, a second simulation is performed on the corrected packaging structure simulation model, simultaneously considering thermal strain and curing shrinkage strain. The second simulation uses the same packaging structure simulation model, the same boundary constraints, and the same temperature history as the first simulation. In the second simulation, in addition to applying thermal strain as in step S210, curing shrinkage strain is also applied to each element of the epoxy molding compound as an initial strain. The curing shrinkage strain is applied according to the curing shrinkage parameters of the epoxy molding compound and is always superimposed on the epoxy molding compound element as an initial strain term in each temperature increment step, participating in the solution of the finite element equilibrium equation. After the second simulation is completed, the out-of-plane displacement values ​​at the coordinate positions of each model on the lower surface of the packaging structure are extracted to form the second out-of-plane displacement distribution. Step S230: Subtract the out-of-plane displacement value of the same model coordinate position in the first out-of-plane displacement distribution from the out-of-plane displacement value of each model coordinate position in the second out-of-plane displacement distribution to obtain the difference at each model coordinate position. This difference is the curing shrinkage out-of-plane displacement component caused solely by curing shrinkage strain. Step S240: Extract the out-of-plane displacement values ​​of each node or integration point within the preset reference area of ​​the FC substrate, and use the least squares method to perform plane fitting on the out-of-plane displacement values ​​to minimize the sum of squared residuals between the fitted plane and the out-of-plane displacement values ​​of each point within the preset reference area. Use the fitted plane as the warping reference plane. Step S250: After determining the warping reference plane, the out-of-plane displacement value of each model coordinate position in the second out-of-plane displacement distribution relative to the warping reference plane is taken as the effective out-of-plane displacement of each position, and the difference between the maximum and minimum values ​​of the effective out-of-plane displacement is determined as the initial predicted warping value.

[0027] In this embodiment, after the first and second simulations, the first out-of-plane displacement distribution under the sole action of thermal strain, the second out-of-plane displacement distribution under the combined action of thermal strain and curing shrinkage strain, and the curing shrinkage out-of-plane displacement component distribution obtained from the output of the same simulation model under the same temperature history under two operating conditions are obtained. This processing separates the curing shrinkage deformation component by setting a control condition, so that the contributions of thermal strain and curing shrinkage to the out-of-plane displacement can be quantified separately. Thus, the stack adjustment can compensate for the out-of-plane deformation direction caused by thermal strain and the out-of-plane deformation direction caused by curing shrinkage respectively, avoiding the deviation in the judgment of the adjustment direction due to the superposition effect of the two deformation factors. At the same time, the initial predicted warpage value serves as a comparison benchmark for the optimization effect of candidate stack designs, which can evaluate whether each candidate stack design has achieved a reduction in warpage.

[0028] In a specific implementation, as one example, step S300 is used to select a material layer on one side of the thickness center plane of the FC substrate for adjustment based on determining the out-of-plane deformation directions caused by thermal strain and curing shrinkage, so that the adjusted stack forms a compensating bending moment opposite to the direction of the molding bending moment. If the stack is adjusted directly based solely on the total deformation distribution, it is impossible to distinguish the respective directions of action of the thermally induced bending moment and the curing shrinkage bending moment, which may result in the adjusted stack compensating for one deformation factor while aggravating the other. This embodiment determines the total curvature components and their directions in each direction, and then selects the material layer to be adjusted based on these directions and implements the adjustment, as follows: Step S310: Determine a first direction and a second direction that are perpendicular to each other in a plane parallel to the FC substrate; the selection of the first direction and the second direction corresponds to the geometry of the FC substrate. When the FC substrate is rectangular, the first direction is parallel to one side of the rectangle and the second direction is parallel to the other side of the rectangle. Step S320: Based on the first out-of-plane displacement distribution, determine the thermal strain curvature components in the first direction and the second direction respectively; for the first direction, select multiple equally spaced sampling points located on the lower surface of the FC substrate in the first out-of-plane displacement distribution along the direction, and perform quadratic curve fitting with the coordinate values ​​of each sampling point in the direction as the independent variable and the out-of-plane displacement value as the dependent variable, and take twice the coefficient of the quadratic term of the fitted curve as the thermal strain curvature component in the direction; for the second direction, select equally spaced sampling points along the second direction in the same way and perform quadratic curve fitting, and take twice the coefficient of the quadratic term as the thermal strain curvature component in the second direction; Step S330: Based on the out-of-plane displacement component of curing shrinkage, determine the curing shrinkage curvature components in the first direction and the second direction respectively; for the first direction, select multiple equally spaced sampling points located on the lower surface of the FC substrate in the distribution of the out-of-plane displacement component of curing shrinkage along the direction, and perform quadratic curve fitting with the coordinate value of each sampling point in the direction as the independent variable and the out-of-plane displacement component of curing shrinkage as the dependent variable, and take twice the coefficient of the quadratic term of the fitted curve as the curing shrinkage curvature component in the direction; for the second direction, select equally spaced sampling points along the second direction in the same way and perform quadratic curve fitting, and take twice the coefficient of the quadratic term as the curing shrinkage curvature component in the second direction. Step S340: Add the thermal strain curvature component and the curvature component of curvature in the same direction to obtain the total curvature component in the corresponding direction; the sign of the curvature component indicates the concave-convex direction of the out-of-plane displacement curve, which corresponds to the bending direction of the FC substrate caused by the molding bending moment in that direction; when the total curvature component is positive, the molding bending moment acts in the positive direction of that direction; when the total curvature component is negative, the molding bending moment acts in the negative direction of that direction. Step S350: Each material layer located on both sides of the thickness center plane of the FC substrate and not crossing the thickness center plane is identified as a test material layer; the thickness center plane of the FC substrate refers to a plane located at half the total thickness of the FC substrate and parallel to the surface of the FC substrate; the material layers located on both sides of the thickness center plane and not crossing the thickness center plane include each material layer located above the thickness center plane and each material layer located below the thickness center plane; the conductor layer, dielectric layer, and solder mask layer are all included as material layers in the determination of the test material layers; for composite layers or sub-layer structures containing multiple materials, if the entire sub-layer structure is located on one side of the thickness center plane and does not cross the thickness center plane, then the sub-layer structure is treated as a material layer; Step S360: For each test material layer, while keeping the material type and layer thickness of all other material layers in the FC substrate consistent with the initial stack-up data, change the material type or layer thickness of the test material layer to form a test stack-up design; the number of conductor layers, the position of the connection pads, and the electrical connection relationship of the FC substrate in the test stack-up design are the same as the initial stack-up data. Step S370: Using the corrected packaging structure simulation model, perform packaging deformation simulation on each trial stack design under the same molding process conditions to obtain the candidate out-of-plane displacement distribution corresponding to each trial stack design; the candidate out-of-plane displacement distribution includes the out-of-plane displacement value at each model coordinate position. Step S380: For each candidate out-of-plane displacement distribution of a test stack design, determine the thermal strain curvature components and curing shrinkage curvature components in the first and second directions corresponding to the candidate out-of-plane displacement distribution, respectively. Then, add the thermal strain curvature components and curing shrinkage curvature components in the same direction to obtain the total curvature components in the first and second directions corresponding to the test stack design, and then determine the predicted curvature direction. Compare the predicted curvature direction with the total curvature direction. If the predicted curvature direction of a certain test stack design is opposite to the direction of the molding moment, then the test material layer that was adjusted in the test stack design is determined as the material layer to be adjusted. When there are multiple test material layers whose predicted curvature direction is opposite to the direction of the molding moment, the test material layer with the largest angle between the predicted curvature direction and the direction of the molding moment or the most significant change in the predicted curvature amplitude is determined as the material layer to be adjusted. Step S390: Adjust at least one of the material type and layer thickness of the material layer to be adjusted so that the adjusted stack forms a compensating bending moment opposite to the direction of the molding bending moment, thus obtaining a candidate stack design; when forming the candidate stack design, keep the number of conductor layers, the position of the connection pads, and the electrical connection relationship of the FC substrate unchanged; the line impedance corresponding to the candidate stack design is within the preset impedance range. The line impedance is determined according to the transmission line impedance calculation formula based on the line width, line spacing, and the thickness and dielectric constant of the adjacent dielectric layers in the candidate stack design. The preset impedance range is set according to the electrical interface requirements of the chip and the substrate.

[0029] In one material type adjustment method, the thickness of the material layer to be adjusted is kept constant, and the material layer to be adjusted is replaced with a candidate material; at least one of the elastic modulus and coefficient of thermal expansion of the candidate material is different from the corresponding parameter of the material layer to be adjusted in the initial stacking data; the elastic modulus of the candidate material is higher or lower than the elastic modulus of the raw material, or the coefficient of thermal expansion of the candidate material is higher or lower than the coefficient of thermal expansion of the raw material, or both are changed simultaneously; by changing the elastic modulus or coefficient of thermal expansion of the material layer to be adjusted, the magnitude or direction of the stress generated by the layer under the same temperature change and curing shrinkage is changed, thereby causing the adjusted stack to form a compensating bending moment opposite to the molding bending moment in the corresponding direction.

[0030] In one layer thickness adjustment method, the material type of the material layer to be adjusted is kept constant, the layer thickness of the material layer to be adjusted is changed, and another material layer in the FC substrate is adjusted by a layer thickness change that is equal in magnitude and opposite in direction to the layer thickness change of the material layer to be adjusted, so that the total thickness of the substrate of the candidate stack design is the same as the total thickness of the substrate corresponding to the initial stack data; for example, if the layer thickness of the material layer to be adjusted is increased by Δh, then Δh is subtracted from the layer thickness of the other material layer; if the layer thickness of the material layer to be adjusted is decreased by Δh, then Δh is added to the layer thickness of the other material layer; the other material layer is selected as the material layer in the FC substrate located on the side opposite to the thickness center plane of the material layer to be adjusted, so that the total thickness adjustment keeps the total thickness of the substrate constant.

[0031] After the above processing, the thermal strain curvature component and the curvature component of curvature shrinkage are obtained respectively. The two are superimposed to obtain the total curvature component and its direction. Then, the material layer adjustment scheme is determined by the encapsulation deformation simulation of the trial stack design. The final candidate stack design can form a reverse compensation bending moment in the direction of the molding bending moment while keeping the number of conductor layers, the position of the pads and the electrical connection relationship unchanged, thereby reducing the possibility of warping aggravation due to improper stack adjustment direction.

[0032] In a specific implementation, as one example, step S400 is used to perform encapsulation deformation simulation on each candidate stack design. Based on the comparison between the predicted warpage value and the initial predicted warpage value of each candidate stack design, the candidate stack design with improved warpage is determined as the target stack design. If the simulation results of a single candidate stack design are used for judgment, it is impossible to determine the degree of improvement of the candidate design relative to the initial design, and there is also a lack of a convergence mechanism to gradually approach the optimal stack parameters in multiple iterations. This embodiment optimizes the stack design and the encapsulation material in two dimensions by setting an iterative optimization process and epoxy molding compound matching and screening, as follows: Step S410: Perform encapsulation deformation simulation on each candidate stack-up design under molding process conditions. These molding process conditions are the same as those used in the first and second simulations in step S200, including the same temperature history, boundary constraints, and curing shrinkage strain application method. Input each candidate stack-up design into the corrected encapsulation structure simulation model, replace the corresponding parameters of each material layer of the FC substrate in the model, and perform encapsulation deformation simulation on each candidate stack-up design while keeping the chip and epoxy molding parameters unchanged, considering both thermal strain and curing shrinkage strain. After the encapsulation deformation simulation of each candidate stack-up design is completed, extract the out-of-plane displacement values ​​at each model coordinate position on the lower surface of the FC substrate in each candidate stack-up design to form the candidate out-of-plane displacement distribution corresponding to each candidate stack-up design. Step S420: Fit a warpage reference plane based on a preset reference region of the FC substrate; the preset reference region is the area on the lower surface of the FC substrate where out-of-plane displacement is constrained or deformation is minimal during the packaging process; for each candidate stack-up design, extract the out-of-plane displacement values ​​of each node or integration point within the preset reference region, and use the least squares method to fit the out-of-plane displacement values ​​to a plane, using the fitted plane as the warpage reference plane corresponding to the candidate stack-up design; take the out-of-plane displacement values ​​of each model coordinate position in the candidate out-of-plane displacement distribution relative to the warpage reference plane as the effective out-of-plane displacement at each position, and determine the difference between the maximum and minimum values ​​of the effective out-of-plane displacements as the predicted warpage value of the candidate stack-up design, and determine the predicted warpage value corresponding to each candidate stack-up design one by one in the above manner; Step S430: Compare the predicted warpage values ​​of each candidate stack design with the initial predicted warpage values ​​one by one; for any candidate stack design, if its predicted warpage value is less than the initial predicted warpage value, then the candidate stack design is determined as the target stack design; if there are multiple candidate stack designs whose predicted warpage values ​​are all less than the initial predicted warpage value, then the candidate stack design with the smallest predicted warpage value is determined as the target stack design; if the predicted warpage values ​​of all candidate stack designs are not less than the initial predicted warpage value, it indicates that no effective candidate scheme to reduce warpage has been formed in the current stack parameter space. At this time, the initial stack design is taken as the target stack design, and adjustment suggestions are output or the current iteration ends. Step S440: Use the target stack design determined in the current design iteration as the initial stack data for the next design iteration, and repeat the determination of the material layer to be adjusted, the formation of candidate stack designs, and the packaging deformation simulation until the predicted warpage value corresponding to the target stack design is not greater than the preset warpage threshold. Then, determine the current target stack design as the final stack design. The preset warpage threshold is set according to the subsequent assembly process requirements of the packaging structure, such as the electrical connection reliability requirements between the chip and the substrate and the interface bonding strength requirements between the molding compound and the substrate. Step S450: Obtain the thermomechanical parameters and curing shrinkage parameters corresponding to multiple candidate epoxy molding compounds. The curing shrinkage parameters of each candidate epoxy molding compound are obtained in the same way as in step S100. Step S460: Use the final stack-up design as the fixed stack-up design and replace the FC substrate stack-up parameters in the packaging structure simulation model; for each candidate epoxy molding compound obtained, replace the epoxy molding compound material parameters in the packaging structure simulation model with its thermomechanical parameters and curing shrinkage parameters, and perform packaging deformation simulations respectively under the condition of keeping the molding process unchanged to obtain the candidate out-of-plane displacement distribution corresponding to each candidate epoxy molding compound; determine the predicted warpage value corresponding to each candidate epoxy molding compound according to the method of step S420; compare the predicted warpage values ​​corresponding to all candidate epoxy molding compounds, and determine the candidate epoxy molding compound with the smallest predicted warpage value as the target epoxy molding compound that matches the target stack-up design.

[0033] In this embodiment, the stack parameters are updated round by round in the direction of warp reduction through iterative optimization until the predicted warp value meets the preset warp threshold. After the stack design converges, a target epoxy molding compound that matches the stack design is further screened. Thus, the stack design and molding compound are optimized in two dimensions respectively, so that the final combination of the target stack design and the target epoxy molding compound has the minimum predicted warp value under the preset molding process conditions, reducing the risk of warp caused by material and stack mismatch.

[0034] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for optimizing warpage during molding of an FC substrate, characterized in that, include: Acquire the initial stack-up data of the FC substrate, the encapsulation structure data, and the thermomechanical parameters and curing shrinkage parameters of the epoxy molding compound, and establish a simulation model of the encapsulation structure including the FC substrate, the chip, and the epoxy molding compound; Under the same molding process conditions, the simulation model of the packaging structure is subjected to a first simulation that only considers thermal strain and a second simulation that simultaneously considers thermal strain and curing shrinkage strain to obtain the first and second out-of-plane displacement distributions. The curing shrinkage out-of-plane displacement component is determined based on the difference between the two, and the initial predicted warpage value is determined based on the second out-of-plane displacement distribution. The direction of the molding bending moment is determined based on the first out-of-plane displacement distribution and the out-of-plane displacement component of the curing shrinkage. Based on this, the material layer to be adjusted on one side of the thickness center plane of the FC substrate is determined. At least one of the material type and layer thickness of the material layer to be adjusted is adjusted so that the adjusted stack forms a compensating bending moment opposite to the direction of the molding bending moment, thus obtaining a candidate stack design. Under the molding process conditions, the candidate stack design is subjected to encapsulation deformation simulation, and the candidate stack design with a predicted warpage value less than the initial predicted warpage value is determined as the target stack design.

2. The FC substrate molding warpage optimization method according to claim 1, characterized in that, The initial stack-up data includes the stack-up order, material type, layer thickness, and conductor pattern distribution of each conductor layer, dielectric layer, and solder resist layer in the FC substrate. The packaging structure data includes the geometric dimensions and relative positions of the FC substrate, the chip, and the epoxy encapsulation.

3. The FC substrate molding warpage optimization method according to claim 1, characterized in that, Before performing the first and second simulations, the measured warpage value of the encapsulation structure formed with a known stack design and a known epoxy molding compound under known molding process conditions is obtained, and the encapsulation structure simulation model is used to perform simulations including thermal strain and curing shrinkage strain to obtain the corresponding simulated warpage value. Based on the absolute value of the difference between the measured warpage value and the simulated warpage value, at least one of the material parameters and interface constraint parameters of the packaging structure simulation model is adjusted until the absolute value is not greater than a preset correction threshold.

4. The FC substrate molding warpage optimization method according to claim 1, characterized in that, Obtain the thermomechanical parameters and curing shrinkage parameters of multiple candidate epoxy molding compounds; While keeping the target stack design and molding process conditions unchanged, the encapsulation deformation simulation is performed using the thermomechanical parameters and curing shrinkage parameters corresponding to each candidate epoxy molding compound, and the candidate epoxy molding compound with the smallest predicted warpage value is determined as the target epoxy molding compound that matches the target stack design.

5. The FC substrate molding warpage optimization method according to claim 1, characterized in that, The target stack design determined in the current design iteration is used as the initial stack data for the next design iteration. The determination of the material layer to be adjusted, the formation of the candidate stack design, and the encapsulation deformation simulation are repeated until the predicted warpage value corresponding to the target stack design is not greater than the preset warpage threshold.

6. The FC substrate molding warpage optimization method according to claim 1, characterized in that, The first simulation and the second simulation use the same packaging structure, boundary constraints, and temperature history; In the first simulation, the curing shrinkage strain is not applied. In the second simulation, the curing shrinkage strain is applied, and the out-of-plane displacement value of each model coordinate position in the second out-of-plane displacement distribution is subtracted from the out-of-plane displacement value of the same model coordinate position in the first out-of-plane displacement distribution to obtain the curing shrinkage out-of-plane displacement component.

7. The FC substrate molding warpage optimization method according to claim 6, characterized in that, A first direction and a second direction that are perpendicular to each other are determined in a plane parallel to the FC substrate; Based on the first out-of-plane displacement distribution, the thermal strain curvature components in the first direction and the second direction are determined respectively; based on the curing shrinkage out-of-plane displacement components, the curing shrinkage curvature components in the first direction and the second direction are determined respectively. The thermal strain curvature component and the curing shrinkage curvature component in the same direction are added together to obtain the total curvature component in the corresponding direction, and the direction of action of the molding bending moment in the corresponding direction is determined according to the positive and negative directions of the total curvature component.

8. The FC substrate molding warpage optimization method according to claim 7, characterized in that, Each material layer located on both sides of the thickness center plane of the FC substrate and not crossing the thickness center plane is defined as a test material layer; For each of the aforementioned test material layers, a test stack design is formed that adjusts only at least one of the material type and layer thickness of the test material layer, and the test stack design is subjected to encapsulation deformation simulation. Based on the predicted curvature direction corresponding to the trial layer design, the trial material layer whose predicted curvature direction is opposite to the direction of the molding bending moment is determined as the material layer to be adjusted.

9. The FC substrate molding warpage optimization method according to claim 8, characterized in that, While keeping the thickness of the material layer to be adjusted constant, the material layer to be adjusted is replaced with a candidate material, wherein at least one of the elastic modulus and coefficient of thermal expansion of the candidate material is different from the corresponding parameter of the material layer to be adjusted in the initial stacking data, so that the candidate stacking design using the candidate material forms the compensating bending moment.

10. The FC substrate molding warpage optimization method according to claim 8, characterized in that, While keeping the material type of the material layer to be adjusted unchanged, the thickness of the material layer to be adjusted is changed, and another material layer in the FC substrate is adjusted by a thickness change that is equal in magnitude and opposite in direction to the thickness change of the material layer to be adjusted, so that the total thickness of the substrate of the candidate stack design is the same as the total thickness of the substrate corresponding to the initial stack data.

Images