H. Pape

Delamination and combined compound cracking of EMC-copper interfaces

The present study deals with experimental investigation of the delamination toughness of EMC (epoxy molding compound) and Copper-leadframe interfaces. Test samples were directly obtained from the production line. EMC is attached on copper substrates with various surface treatments. Mixed mode bending experiments were performed under various temperature and moisture environments. The test procedure and some results were reported previously in ECTC08 and ECTC09 [1–2]. Recently, we studied the effect of delaminated surfaces in order to get better understanding of the established fracture toughness. Therefore, after the delamination experiments, some of the delaminated samples were subjected to various surface analyses (SEM, FIB, EDX). Two types of failure patterns are found depending on the loading mode mixture, and the environmental conditions. Firstly, depending on the type of copper surface treatment, pure interface delamination is observed for some of the interfaces. Here, we observed clean delaminated copper surfaces. The second type of failure is a combination of interface delamination and compound cracking. Here, it is found that after the separation of interfaces, some EMC remains on the copper surface. In this case the experiment results showed that the interface delamination and molding compound cracking combined failure occurs at relatively high force values.

Interfacial fracture parameters of silicon-to-molding compound

The rapid diversification in microelectronics forebodes more complex system integration, be it for denser function integration or a span of dimensions between various technologies. Products may include more features, perform faster and be cheaper. With these trends the amount of material layers is increasing. This challenges development to a faster rating of material pairings. Delamination is a major issue among the related reliability aspects. When the design or testing steps are accompanied by simulation, fracture mechanical descriptions are increasingly proving helpful. The parameters needed for simulation have to be measured and should be available for different fracture mode mix angles. We investigated the interfacial fracture toughness of the Epoxy Molding Compound (EMC) to Silicon interface. Although difficult to delaminate we could carry out measurements using the Mixed Mode Chisel setup (MMC) that allowed us to induce different stress states at the crack tip at various external load angles. The samples we derived from the molding process of embedded wafer level ball grid arrays. Therefore we were able to use samples made with the same process as in real packaging. The crack tip position was determined by analysis of displacement results by digital image correlation. In order to interpret the sample reaction for extracting fracture mechanical parameters, adequate numerical modeling and simulation was required. The experiments provided the parameters for the models. Establishing the residual stress state in the materials preceded the interface delamination simulation: a two step interpretation. Residual stresses cannot be neglected; indeed they are part of the challenges to delaminate this interface at all. We found energy release rates increasing with fracture mode mix, and such values close to pure tensile opening at the crack tip. We recommend to exclude data from short crack lengths and to carefully expose the sample flanks. The results promise to extend the available interfacial fracture data soon.

Interface fracture mechanics evaluation by correlation of experiment and simulation

Interface fracture mechanics is one of the main focuses of electronics reliability research. Determination of fracture mechanical properties of interface cracks is a substantial task for design for reliability concept. Without experimental determined fracture mechanical parameters such as the critical energy release rate a reliability forecast based on simulation results cannot be given. In fracture mechanics testing often a correct measurement of the crack tip location is needed for the calculation of the energy release rate. The authors present a combined simulative and experimental method for crack tip location determination of typical interface specimens. The specimens are loaded in a newly designed testing apparatus, the Mixed Mode Chisel (MMC) setup, and images of the crack tip at the interface are taken at different load states during the testing procedure. Then images are analyzed by image correlation techniques (DAC, deformation analysis by correlation) and crack tip displacement fields are determined. In the next analysis step the displacement fields are compared to fields from finite element analysis of the same specimen geometry with boundary conditions similar to the experimental setup. The point of the best matching of the experimental and simulative field is the actual crack tip location. If finite-element data or analytical solution for the crack tip displacement field is available the method can be applied for a variety of different interface samples.

Establishing fracture properties of EMC-Copper interfaces in the visco-elastic temperature region

An ongoing root cause of failure in microelectronic industry is interface delamination. In order to explore the risk of interface damage, FE simulations for the fabrication steps as well as for the testing conditions are generally made in the design stage. In order to be able to judge the risk for interface fracture, the critical fracture properties of the interfaces being applied should be available, for the occurring combinations of temperature and moisture preconditioning. As a consequence there is an urgent need to establish these critical interface fracture parameters. For brittle interfaces such as between epoxy molding compound (EMC) and metal (-oxide) substrates the critical energy release rate (or delamination toughness) can be considered as the suitable material parameter. This material parameter is strongly dependent on the temperature, the moisture content of the materials involved and on the so-called mode-mixity of the stress state near the crack tip. The present study deals with experimental investigation of the delamination toughness of EMC-Copper lead-frame interfaces as can directly be obtained from the production line. The experimental set-up as designed for this purpose was previously reported, together with some measurement results and toughness evaluations for room temperature fracture tests. This study deals with experiment and simulation procedure of establishing the interfacial fracture toughness from fracture test results at high temperatures, especially in the glass transition temperature region of epoxy molding compound (EMC). In order to calculation accurate fracture toughness, the material property of molding compound is characterized as a function of temperature. A detailed discussion of how EMC responses at its glass transition region will be provided. The influence of the material property on interfacial fracture toughness will be given.

Establishing mixed mode fracture properties of EMC-Copper (-oxide) interfaces at various temperatures

Interfacial delamination is known as one of the root causes of failure in microelectronic industry. In order to explore the risk of interface damage, FE simulations for the fabrication steps as well as for the testing conditions are generally made in the design stage. In order to be able to judge the risk for interface fracture, the critical fracture properties of the interfaces being applied should be available, for the occurring combinations of temperature and moisture preconditioning. As a consequence there is an urgent need to establish these critical interface fracture parameters. For brittle interfaces such as between epoxy molding compound (EMC) and metal (-oxide) substrates the critical energy release rate (or delamination toughness, Gc) can be considered as the suitable material parameter. This material parameter is strongly dependent on the temperature, the moisture content of the materials involved and on the so-called mode mixity of the stress state near the crack tip. The present study deals with experimental investigation of the delamination toughness of EMC-Copper lead-frame interfaces as can directly be obtained from the production line. The experimental set-up as designed for this purpose was previously reported [1], together with some measurement results and toughness evaluations for room temperature fracture tests. This study deals with the experimental and simulation procedures to establish the interfacial fracture toughness from fracture test results at different temperatures, especially in the glass transition temperature region of epoxy molding compound. In order to calculate accurate fracture toughness, the viscoelastic material properties of molding compound are measured and considered. A special test procedure used to investigate the fracture properties in the glass transition temperature region of EMC will be introduced. The FE model used to simulate the viscoelastic material behavior will be discussed. The delamination toughness as a function of mode mixity at different temperatures will be given in the result section.

Establishing mode mix dependency of fracture toughness in microelectronic components with reduced experimental effort

Bulk material fracture and interface delamination are main failure modes observed in microelectronic components. Drivers are stresses during processing (e.g. soldering), testing (e.g. moisture sensitivity, thermal cycling) and operation in application environments. For quantitative modeling to predict failures the fracture toughness expressed as critical energy release rate Gc has to be known by measurements. As interface toughness depends on processing of materials, the mode mix of tensile and shear stress loads expressed by a mode angle Ψ with a reference length lref as additional parameter, further on temperature T and moisture condition C, this is a huge task and full data sets are rarely found. In summary Gc=Gc(Ψ(lref), T, C) is a function of four parameters. Usually just fit functions of measured data are reported. A physics based relation of the dependency of critical energy release rate Gc on mode angle Ψ is highly desirable, could drastically reduce the experimental effort, and provide a model to better understand influence factors. The present work suggests such a relation. Starting from an idealized layer stack of inert particles a hypothesis is derived, which leads to a basic relation for Gc(Ψ). We apply this function to four cases to fit own experimental results and data from literature or phenomenological expressions representing measured data. Good agreement is found. Then we extend the approach from an idealized flat interface to more realistic rough interfaces to explain adhesion promotion by roughening for asymmetric and symmetric roughness profiles. Finally we suggest an expression to cover also the effect of chemical adhesion promotion. The expressions derived cover all aspects of adhesion and depend on 1-3 parameters only. In principal a similar number of independent measurements at different mode angles Ψ is enough to establish the whole Gc(Ψ) curve. More is safer. We suggest to do about five measurements at suitable mode angles Ψ. That should give all information needed. With more experience, and for extensions to different temperature and humidity conditions, three or in case of poor adhesion even one additional experiment can be sufficient. Overall the approach enables a tremendous reduction of the experimental characterization effort and a better understanding of interface adhesion.

Establishing Fracture Properties of EMC-Copper (-Oxide) Interfaces: Test Procedures and Simulations for Establishing the Interface Toughness, Depending on Temperature, Humidity and Mode Mixity

Interfacial delamination has become one of the key reliability issues in the microelectronics of portable devices and therefore is getting more and more attention. The analysis of delamination of a laminate structure with a crack along the interface is central to the characterization of interfacial toughness. Due to the mismatch in mechanical properties of the materials adjacent to the interface and also possible asymmetry of loading and geometry, usually the delamination propagates under mixed mode conditions. In this study, a modified mixed mode bending test using production line interface samples is proposed. The critical fracture properties are obtained by interpreting the experimental results through dedicated finite element modeling. The interface types being considered in the present work are between EMCs and copper lead frame.

Mixed mode interface characterization considering thermal residual stress

Interfacial delamination has become one of the key reliability issues in the microelectronic industry and therefore is getting more and more attention. The analysis of delamination of a laminate structure with a crack along the interface is central to the characterization of interfacial toughness. Due to the mismatch in mechanical properties of the materials adjacent to the interface and also possible asymmetry of loading and geometry, usually the crack propagates under mixed mode conditions. The present study deals with delamination toughness measurements of an epoxy molding compound - copper lead frame interface as directly obtained from a real production process. As a consequence the specimen dimensions are relatively small and therefore a dedicated small-size test set-up was designed and fabricated. The test setup allows transferring two separated loadings (mode I and mode II) on a single specimen. The setup is flexible and adjustable for measuring specimens with various dimensions. For measurements under various temperatures and moisture conditions, a special climate chamber is designed. The ldquocurrent crack lengthrdquo is required for the interpretation of measurement results through FEM-fracture mechanics simulations. Therefore, during testing the ldquocurrent crack lengthrdquo is captured using a CCD camera and a micro deformation analysis system (MicroDac). The critical fracture properties are obtained by interpreting the experimental results through dedicated finite element modeling.

Fracture-mechanical interface characterisation for thermo-mechanical co-design — An effcient and comprehensive method for critical mixed-mode data extraction

This paper presents a comprehensive method for obtaining urgently required critical interface delamination data of material pairings used in electronic packaging. The objective is to thereby enable rapid, inexpensive and accurate lifetime prediction for that failure mode. A new testing method is presented which allows large mode-angle range and enhanced throughput testing under multiple loading conditions, the coverageof which is usually a rather lengthy and resource-demanding procedure. The approach is specimen-centred in the sense that the accent is put on test-specimens which are easily manufacturable industrially, rather than having to adapt them to a special testing machine. The concept is also scalable, i.e. it has potential to work also for smaller samples cut from real devices. We show the first version of a newly developed test-stand and discuss first results for copper-molding compound interfaces and interpret them using simulations.

Delamination toughness of Cu-EMC interfaces at harsh environment

Interfacial delamination has become one of the key reliability issues in the microelectronic industry and therefore is getting more and more attention. The analysis of delamination of a laminate structure with a crack along the interface is central to the characterization of the interfacial toughness. Interfacial toughness is highly dependent to temperature, moisture and mode mixity. The present study deals with delamination toughness measurements of an epoxy molding compound — copper lead frame interface as directly obtained from a real production process in harsh environment, Temperature >100 °C & 100% RH. To deal with it, a chamber with high pressure, i.e. pressure vessel or pressure cooker, is needed. Controlling the inside pressure makes it possible to have 100% RH at different temperature levels. In addition to the initial stress state due to the harsh environment, mechanical loading under combined mode I/ II conditions is applied on a bi-material specimen to initiate and propagate the delamination. For this a mixed mode bending setup is installed in the pressure chamber.