CN-122021113-A - Method and device for constructing rate-dependent cohesive force model based on interface degradation state

CN122021113ACN 122021113 ACN122021113 ACN 122021113ACN-122021113-A

Abstract

The invention relates to the technical field of electronic packaging, and discloses a method and a device for constructing a rate-related cohesion model based on an interface degradation state, wherein the method comprises the steps of carrying out a temperature cycle experiment on a plurality of wafer-level package RDL samples in the same batch to enable a PI-Cu interface to be in the degradation state, carrying out an interface thrust experiment at a plurality of different push-out knife speeds, and obtaining force-displacement experiment curves corresponding to the different push-out knife speeds; obtaining parameters of a cohesive force model of a PI-Cu interface under different push broach rates through finite element simulation inversion, wherein the parameters comprise interface rigidity, interface shear strength and critical energy release rate, fitting a functional relation between the parameters of the cohesive force model and the push broach rates, and constructing a rate-related cohesive force model. The method solves the problem of insufficient characterization of mechanical properties of the rate-related interface in the degradation state in the prior art, provides an accurate theoretical basis for evaluating and optimizing the design of the reliability of the wafer-level package, and effectively improves the reliability level of the packaging technology.

Inventors

FU HUALONG
ZHAO JINGYI
SU MEIYING
MA RUI
CHEN CHUAN
ZHAO QUANLU
LI JUN
WANG QIDONG

Assignees

中国科学院微电子研究所

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (10)

1. The method for constructing the rate-dependent cohesive force model based on the interface degradation state is characterized by comprising the following steps of: Carrying out a temperature cycling experiment of a preset temperature variation range and a preset cycle number on a plurality of wafer-level package RDL samples in the same batch, so that PI-Cu interfaces of the RDL samples are in a degradation state; carrying out interface thrust experiments on the RDL sample subjected to the temperature cycle experiment at a plurality of different push-broach speeds to obtain force-displacement experiment curves corresponding to the different push-broach speeds; acquiring cohesive force model parameters of the PI-Cu interface at different push broach speeds through finite element simulation inversion based on the force-displacement experimental curve, wherein the cohesive force model parameters comprise interface rigidity, interface shear strength and critical energy release rate; And fitting a functional relation between the cohesive force model parameter and the push broach speed, and constructing a rate-dependent cohesive force model for representing the mechanical behavior of the PI-Cu interface in the degradation state.
2. The method of claim 1, wherein after performing a predetermined temperature range and a predetermined number of temperature cycling experiments on a plurality of wafer level package RDL samples of a same lot, the method further comprises: Observing the PI-Cu interface morphology of the RDL sample by using an ultrasonic scanning microscope to obtain an observation result; and evaluating the degradation degree of the PI-Cu interface based on the observation result, and confirming that the PI-Cu interface reaches a preset degradation state according to the degradation degree.
3. The method of claim 1, wherein the obtaining cohesive model parameters of the PI-Cu interface at different push-broach rates based on the force-displacement experimental plot and by finite element simulation inversion comprises: calculating to obtain a critical energy release rate according to the force-displacement experimental curve; Constructing a three-dimensional finite element model based on a PI-Cu-Si three-layer structure of the RDL sample, setting cohesive force contact attribute on the PI-Cu interface, adopting a maximum nominal stress criterion as an initial damage rule, selecting an energy-based damage evolution rule, and setting interface rigidity and interface shear strength; simulating the load and boundary conditions of an interface thrust experiment in the three-dimensional finite element model, and calculating to obtain a force-displacement simulation curve; And adjusting the interface rigidity and the interface shear strength to enable the force-displacement simulation curve to be matched with the force-displacement experimental curve, and determining cohesive force model parameters based on the critical energy release rate, the adjusted target interface rigidity and the adjusted interface shear strength.
4. A method according to claim 3, wherein the critical energy release rate is calculated as: wherein: is critical energy release rate, P is the force value in the force-displacement experimental curve, A is the contact area of the PI-Cu interface, Is a infinitesimal of the displacement, And the area from the starting point to the PI-Cu interface complete failure point in the force-displacement experiment curve is shown.
5. A method according to claim 3, characterized in that the method further comprises: monitoring the stiffness reduction rate of the PI-Cu interface, wherein the stiffness reduction rate is the difference of the ratio of the current interface stiffness to the initial interface stiffness of the PI-Cu interface; and when the rigidity reduction rate is 1, judging that the PI-Cu interface is completely failed.
6. The method of claim 3, wherein prior to said determining cohesion model parameters based on said critical energy release rate, said target interface stiffness after adjustment, and said interface shear strength after adjustment, said method further comprises: Selecting a target pushing rate, counting cohesion model parameters obtained by a plurality of RDL samples of the same batch under the target pushing rate, calculating an arithmetic average value of the cohesion model parameters, and taking the arithmetic average value as a final cohesion model parameter under the target pushing rate.
7. The method of claim 1, wherein the functional relationship between the cohesion model parameter and the push rate is as follows: wherein: Is the critical energy release rate; Is the critical energy release rate under the reference condition; K is the interface rigidity; Is the interfacial stiffness under reference conditions; Is interfacial shear strength; The shear strength of the interface under the reference condition, v is the push-broach speed, The temperature variation range of the temperature cycle experiment is defined, and n is the cycle number of the temperature cycle experiment.
8. A rate-dependent cohesion model construction apparatus based on an interface degradation state, comprising: The interface degradation module is used for carrying out temperature cycle experiments of a preset temperature range and a preset number of times on a plurality of wafer-level package RDL samples in the same batch, so that PI-Cu interfaces of the RDL samples are in a degradation state; The curve generation module is used for carrying out interface thrust experiments on the RDL sample subjected to the temperature cycle experiment at a plurality of different push-broach speeds to obtain force-displacement experiment curves corresponding to the different push-broach speeds; The parameter acquisition module is used for acquiring cohesive force model parameters of the PI-Cu interface under different push broach rates through finite element simulation inversion based on the force-displacement experimental curve, wherein the cohesive force model parameters comprise interface rigidity, interface shear strength and critical energy release rate; the model construction module is used for fitting the functional relation between the cohesion model parameters and the push broach speed and constructing a rate-related cohesion model for representing the mechanical behavior of the PI-Cu interface in the degradation state.
9. A storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the method of any of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program when executed by the processor implements the steps of the method according to any one of claims 1 to 7.

Description

Method and device for constructing rate-dependent cohesive force model based on interface degradation state Technical Field The invention relates to the technical field of electronic packaging, in particular to a method and a device for constructing a rate-dependent cohesive force model based on an interface degradation state. Background In Wafer-level packaging (Wafer-LEVEL PACKAGING, WLP) technology, a redistribution layer (Redistribution Layer, RDL) is used as a core structure for realizing electrical interconnection between an internal I/O pad and an external package pin of a chip, and is widely applied to advanced packaging scenes such as 2.5D/3D multi-chip integration, and the like, and a typical RDL structure is formed by alternately stacking a metal layer (such as copper, cu) and an insulating layer (such as polyimide, PI), wherein the metal layer is used for transmitting power, grounding and high-speed signals, and the insulating layer provides electrical isolation and mechanical support for a metal circuit, and based on the RDL structure, not only the function of external interconnection of the chip is borne, but also high-density and high-speed interconnection among the inter-chip pads is realized in multi-chip stacking, and the reliability directly determines the electrical performance stability and service life of the whole packaging system. However, in the practical application process, interface cracking or delamination failure is very easy to occur due to mismatch of thermal expansion coefficients of materials, process residual stress and external thermal-mechanical load (such as high temperature circulation, drop impact and the like) of polyimide PI and copper Cu interfaces in RDL, so that the current research on the fracture behavior of PI-Cu interfaces is mostly based on constant fracture parameters in ideal states, quantitative characterization of rate-related (i.e. strain rate-dependent) fracture resistance performance of interfaces in degradation states is lacking, and particularly after high-temperature aging or mechanical impact is experienced, the mechanical performance of the interfaces is remarkably evolved, and if reliability simulation is performed by adopting cohesive force model (Cohesive Zone Model, CZM) parameters under static or undegraded conditions, serious misalignment of package failure risk assessment is caused. Therefore, a method capable of combining an actual load test and numerical inversion is needed to obtain the CZM key parameters considering the interface degradation and the push-broach rate coupling effect so as to construct an accurate cohesion model. Disclosure of Invention In view of the above, the application provides a method and a device for constructing a rate-dependent cohesive force model based on an interface degradation state, which mainly aim to solve the technical problem that in the prior art, reliability simulation is performed by adopting cohesive force model parameters under static or undegraded conditions, so that encapsulation failure risk assessment is misaligned. According to a first aspect of the present invention, there is provided a method for constructing a rate-dependent cohesion model based on an interface degradation state, comprising: Carrying out a temperature cycle experiment of a preset temperature variation range and a preset cycle number on a plurality of wafer-level package RDL samples in the same batch, so that PI-Cu interfaces of the RDL samples are formed In a degraded state; carrying out interface thrust experiments on the RDL sample subjected to the temperature cycle experiment at a plurality of different push-broach speeds to obtain force-displacement experiment curves corresponding to the different push-broach speeds; acquiring cohesive force model parameters of the PI-Cu interface at different push broach speeds through finite element simulation inversion based on the force-displacement experimental curve, wherein the cohesive force model parameters comprise interface rigidity, interface shear strength and critical energy release rate; And fitting a functional relation between the cohesive force model parameter and the push broach speed, and constructing a rate-dependent cohesive force model for representing the mechanical behavior of the PI-Cu interface in the degradation state. Optionally, after the temperature cycle experiments of the preset temperature ranges and the preset times are performed on the plurality of wafer-level package RDL samples in the same batch, the method further comprises the steps of observing the PI-Cu interface morphology of the RDL samples by using an ultrasonic scanning microscope to obtain an observation result, evaluating the degradation degree of the PI-Cu interface based on the observation result, and confirming that the PI-Cu interface reaches a preset degradation state according to the degradation degree. The method comprises the steps of obtaining a cohesive force model parameter of