![Volume fraction of IMCs increases as solder joint size decreases [4]](https://www.researchgate.net/profile/Soud-Choudhury/publication/311559659/figure/fig3/AS:562790222254101@1511191053773/olume-fraction-of-IMCs-increases-as-solder-joint-size-decreases-4_Q320.jpg)
Content uploaded by Soud Choudhury
Author content
All content in this area was uploaded by Soud Choudhury on Nov 20, 2017
Content may be subject to copyright.
A preview of the PDF is not available
Intermetallic Compounds and Their Effects on the Mechanical Performance of Micro Scale Solder Bonds
Abstract and Figures
Interconnects in microelectronic packages and devices serve as the mechanical and electrical connections as well as thermal paths for heat dissipation. Miniaturization of electronic devices demands very high density interconnects and solder bonds with only a few microns of stand-off height. Although Intermetallic compounds (IMCs) are essential to form a reliable joint, large volume ratios of IMCs can be degrading to long term reliability. At very small joints, the volume of IMCs becomes significant, and in some cases, joints may completely transform to IMCs. Furthermore, small joints experience anisotropy due to the fact that all compositions may only contain a few grains. However, very few studies have been conducted to analyze the effect of the IMCs thickness and anisotropy on the mechanical behavior of solder bonds. In this work, these effects are studied through a combination of experiments and finite element simulations. Traditionally, regular finite element (FE) modeling techniques have been used to simulate the solder joints. However, in order to evaluate joint with only a few grains, a more sophisticated modeling of elastic and plastic behavior of grains is needed through crystal plasticity finite element (CPFE) modeling. In this study, CPFE is used to model all materials including solder, IMCs, and Cu in joints with different IMC thicknesses. Nanoindentation experiment on single grains of IMCs and CPFE simulation of the same were combined to obtain slip system parameters of IMCs that are necessary constants for CPFE modeling. Furthermore, the electron backscatter diffraction (EBSD) analysis was used to determine the preferred grain growth orientation of Cu6Sn5 IMC on polycrystalline Cu substrate. A lap-shear experiment was designed and conducted to investigate the effects of the different volume fraction of IMCs on the shear behavior of micro-scale solder joints with a 50µm stand-off height. The local strain was measured using a micro-scale Digital Image Correlation (DIC) technique. This experiment was used to determine the local and global stress-strain behavior of these joints. The joints were tested to failure, and fractography was conducted to determine the failure modes and failure sites. Comparison between experiment and modeling shows that the CPFE models are successful in capturing the local mechanical behavior of the solder bonds. CPFE models are observed to be more efficient in predicting the local plastic deformation behavior of micro-scale bonds with few grains than the regular FE analysis. Simulation results show that the overall stress distribution and shear deformation changes as the IMC thickness increases. Stiffer response and higher shear yield strength are seen as the IMC thickness increases for both simulation and experiment results. Also, the stress-strain distributions observed in the CPFE analysis performed to mimic the experiment gave a clear idea of the locations of the possible failure sites. Fractography shows failure mode changing from ductile to brittle where crack propagation path is modulated by the different volume fraction of IMCs. A significant influence of Cu3Sn/Cu6Sn5 interfacial morphology on the ultimate shear strength at a higher volume fraction of IMC samples was observed during the lap-shear experiment where planar interfacial morphology promoted lower ultimate shear strength. The effect of different interfacial morphologies on the shear behavior was verified using CPFE simulation. Calculation of the fatigue life indicates that high volume fractions of IMCs could degrade fatigue life of the solder bonds.
Figures - uploaded by Soud Choudhury
Author content
All figure content in this area was uploaded by Soud Choudhury
Content may be subject to copyright.
... This phenomenon is known as reactive diffusion and has been observed in numerous cases as Cu-Sn (1) , Fe/Sn (2) , Ti-Al (3) , Au-Sn (4) and Ni-Sn (5) . The inter metallic compounds often have properties that are different from the base metals and excessive inter metallic compounds in a couple are detrimental (6) since they alter the mechanical (7) and electrical properties of the joint (8) . Thus, the study of reactive diffusion in binary systems is important for technological as well as for academic purposes. ...
... They reviewed creep-fatigue models in the aerospace and power electronics. Although the investigation by Choudhury et al. [80] identified a new SAC alloy with better thermal and mechanical reliability than Sn-Ag3-Cu0.5/Sn-Ag4-Cu0.5, research has not adequately studied the creep failure. Creep properties at high temperature give an excellent indication of the thermal cycling performance of the solder alloy. ...
Solder joints play a critical role in electronic devices by providing electrical, mechanical and thermal interconnections. These miniature joints are also the weakest links in an electronic device. Under severe thermal and mechanical loadings, solder joints could fail in 'tensile fracture' due to stress overloading, 'fatigue failure' because of the application of cyclical stress and 'creep failure' due to a permanent long-term load. This paper reviews the literature on solder joint failures under thermo-mechanical loading conditions, with a particular emphasis on fatigue and creep failures. Literature reviews mainly focused on commonly used lead-free SnAg -Cu (SAC) solders. Based on literature in experimental and simulation studies on solder joints, it was found that fatigue failures are widely induced by accelerated thermal cycling (ATC). During ATC, the mismatch in coefficients of thermal expansion (CTE) between different elements of electronics assembly contributes significantly to induce thermal stresses on solder joints. The fatigue life of solder joints is predicted based on phenomenological fatigue models that utilise materials properties as inputs. A comparative study of 14 different fatigue life prediction models is presented with their relative advantages, scope and limitations. Creep failures in solder joints, on the other hand, are commonly induced through isothermal ageing. A critical review of various creep models is presented. Many of these strain rate-based creep models are routed to very well-known Anand Model of inelastic strain rate. Finally, the paper outlined the combined effect of creep and fatigue on solder joint failure.
... They reviewed creep-fatigue models in the aerospace and power electronics. Although the investigation by Choudhury et al. [79] identified a new SAC alloy with better thermal and mechanical reliability than Sn-Ag3-Cu0.5/Sn-Ag4-Cu0.5, research has not adequately studied the creep failure. Creep properties at high temperature give an excellent indication of the thermal cycling performance of the solder alloy. ...
Solder joints play a critical role in electronic devices by providing electrical, mechanical and thermal interconnections. These miniature joints are also the weakest links in an electronic device. Under severe thermal and mechanical loadings, solder joints could fail in 'tensile fracture' due to stress overloading, 'fatigue failure' because of the application of cyclical stress and 'creep failure' due to a permanent long-term load. This paper reviews the literature on solder joint failures under thermo-mechanical loading conditions, with a particular emphasis on fatigue and creep failures. Literature reviews mainly focused on commonly used lead-free SnAg -Cu (SAC) solders. Based on literature in experimental and simulation studies on solder joints, it was found that fatigue failures are widely induced by accelerated thermal cycling (ATC). During ATC, the mismatch in coefficients of thermal expansion (CTE) between different elements of electronics assembly contributes significantly to induce thermal stresses on solder joints. The fatigue life of solder joints is predicted based on phenomenological fatigue models that utilise materials properties as inputs. A comparative study of 14 different fatigue life prediction models is presented with their relative advantages, scope and limitations. Creep failures in solder joints, on the other hand, are commonly induced through isothermal ageing. A critical review of various creep models is presented. Many of these strain rate-based creep models are routed to very well-known Anand Model of inelastic strain rate. Finally, the paper outlined the combined effect of creep and fatigue on solder joint failure.
Reliability risk in solder joints has been commonly linked to various failure mechanisms, with electromigration and fatigue among the prominent drivers. Even though these failure mechanisms often occur simultaneously, typical accelerated tests are based on a single failure mechanism. Consequently, the effect of multiple interacting failure mechanisms is ignored, which may underestimate reliability risk. This common practice has the potential to reduce life prediction accuracy. To address this gap, in this paper, we present an experimental set-up that is developed and used to simultaneously evaluate SAC 305 solder joint lifetime under electromigration and fatigue failure mechanisms using sequential and concurrent approaches. The sequential approach involves pre-stressing the solder joint using a fixed current density and temperature until the test vehicle attains a specified rise in resistance. After that, an isothermal fatigue load was applied at three prescribed temperatures. The data collected enables the estimation of parameters used for reliability prediction. On the other hand, the concurrent case involves the simultaneous application of current density, temperature, and cyclic mechanical load to the solder joint. The correlation of the experiment and complementary numerically simulated strain energy density resulted in an improved, physic-based reliability model, which has the added benefit of reduced time and cost of data collection, especially during new product development. Unlike the concurrent approach, the sequential approach shows a defect-driven failure mode. The prediction accuracy of the concurrent-based derived characteristic life model compared to that of experiment-derived characteristic life was found to be 25%.
Low temperature, Sn-based Pb-free solders were developed by making alloy additions to the starting material, 96.5Sn-3.5Ag (mass%). The melting behavior was determined using Differential Scanning Calorimetry (DSC). The solder microstructure was evaluated by optical microscopy and electron probe microanalysis (EPMA). Shear strength measurements, hardness tests, intermetallic compound (IMC) layer growth measurements, and solderability tests were performed on selected alloys. Three promising ternary alloy compositions and respective solidus temperatures were: 91.84Sn-3.33Ag4.83Bi, 212degreesC; 87.5Sn-7.5Au-5.0Bi, 200degreesC; and 86.4Sn-5.1Ag-8.5Au, 205degreesC. A quaternary alloy had the composition 86.8Sn-3.2Ag-5.0Bi-5.0Au and solidus temperature of 194degreesC. The shear strength of this quaternary alloy was nearly twice that of the eutectic Sn-Pb solder. The 66Sn-5.0Ag-10Bi-5.0Au-10In-4.0Cu alloy had a solidus temperature of 178degreesC and good solderability on Cu. The lowest solidus temperature of 159degreesC was realized with the alloy 62Sn-5.0Ag-10Bi-4.0Au-10In-4.0Cu-5.0Ga. The contributing factor towards the melting point depression was the composition of the solid solution, Sn-based matrix phase of each solder.
This study reports on the anisotropic indentation response of a-titanium. Coarse-grained titanium was characterized by electron backscatter diffraction. Sphero-conical nanoindentation was performed for a number of different crystallographic orientations. The grain size was much larger than the size of the indents to ensure quasi-single-crystal indentation. The hexagonal c-axis was determined to be the hardest direction. Surface topographies of several indents were measured by atomic force microscopy. Analysis of the indent surfaces, following Zambaldi and Raabe (Acta Mater. 58(9), 3516-3530), revealed the orientation-dependent pileup behavior of a-titanium during axisymmetric indentation. Corresponding crystal plasticity finite element (CPFE) simulations predicted the pileup patterns with good accuracy. The constitutive parameters of the CPFE model were identified by a nonlinear optimization procedure, and reproducibly converged toward easy activation of prismatic glide systems. The calculated critical resolved shear stresses were 150 6 4, 349 6 10, and 1107 6 39 MPa for prismatic and basal 〈a〉-glide and pyramidal 〈c + a〉-glide, respectively.
This unique book provides an up-to-date overview of the concepts behind lead-free soldering techniques. Readers will find a description of the physical and mechanical properties of lead-free solders, in addition to lead-free electronics and solder alloys. Additional topics covered include the reliability of lead-free soldering, tin whiskering and electromigration, in addition to emerging technologies and research.
Application of the consistent, comprehensive and physically meaningful probabilistic design for reliability (PDfR) concept can not only help to understand the physics-of-failure of an electronic product, but, most importantly, can enable one to predict, quantify and assure its failure-free performance in the field. The use of the PDfR concept can be helpful also in the development and implementation of the new generation of the most feasible and effective qualification test (QT) methodologies, practices and specifications. The major ten PDfR requirements (“commandments”) for the predicted, quantified and assured reliability of an electronic or a photonic product could be formulated as follows:
The nonuniform and localized deformations of ductile single crystals subject to tensile loading are analyzed numerically. The crystal is modelled by a rate independent, elastic-plastic relation based on Schmid's law which precisely accounts for lattice rotations. Both self hardening and latent hardening of the slip systems are included in the model. The crystal geometry is idealized in terms of a planar double slip model. Initial imperfections are specified in the form of slight thickness inhomogeneities and the calculations follow the crystal deformation through diffuse necking and the formation of shear bands. The pattern of shear bands depends on the initial imperfection, but, independent of the particular small imperfection, the material planes of the bands are inclined at a characteristic angle to the slip planes. Also, the lattice misorientation across the shear band, which is such as to cause geometrical softening of the bands, is not sensitive to the imperfection form. For high strength, low hardening crystals a comparison with existing experimental data shows remarkably good qualitative and quantitative agreement between the calculations and observations. We also model a relatively soft high hardening crystal which undergoes more diffuse necking than the strong high hardening crystal. Diffuse necking leads to lattice rotations which produce geometrical softening and hence promote shear band formation. Furthermore, we carry out a calculation for a high strength low hardening crystal with the latent hardening rate prescribed somewhat larger than for isotropic hardening. In this case a 'patchy' pattern of slip emerges. However, the course of shear band development is unaffected.
The effects of void size and hardening in a hexagonal close-packed single crystal containing a cylindrical void loaded by a far-field equibiaxial tensile stress under plane strain conditions are studied. The crystal has three in-plane slip systems oriented at the angle 60° with respect to one another. Finite element simulations are performed using a strain gradient crystal plasticity formulation with an intrinsic length scale parameter in a non-local strain gradient constitutive framework. For a vanishing length scale parameter the non-local formulation reduces to a local crystal plasticity formulation. The stress and deformation fields obtained with a local non-hardening constitutive formulation are compared to those obtained from a local hardening formulation and to those from a non-local formulation. Compared to the case of the non-hardening local constitutive formulation, it is shown that a local theory with hardening has only minor effects on the deformation field around the void, whereas a significant difference is obtained with the non-local constitutive relation. Finally, it is shown that the applied stress state required to activate plastic deformation at the void is up to three times higher for smaller void sizes than for larger void sizes in the non-local material.
The tensile behavior and microstructure of bulk, Sn-3.5Ag solders as a function of cooling rate were studied. Cooling rate
is an important processing parameter that affects the microstructure of the solder and, therefore, significantly influences
mechanical behavior. Controlled cooling rates were obtained by cooling specimens in different media: water, air, and furnace.
Cooling rate significantly affected secondary dendrite-arm size and spacing of the Sn-rich phase, as well as the aspect ratio
of Ag3Sn. The Sn-rich dendrite-arm size and spacing were smaller for water-cooled specimens than for air-cooled specimens. Furnace
cooling yielded a nearly eutectic microstructure because the cooling rate approached equilibrium cooling. The morphology of
Ag3Sn also changed from spherical, at a fast cooling rate, to a needlelike morphology for slower cooling. The changes in the
microstructure induced by the cooling rate significantly affected the mechanical behavior of the solder. Yield strength was
found to increase with increasing cooling rate, although ultimate tensile strength and strain-to-failure seemed unaffected
by cooling rate. Cooling rate did not seem to affect Young’s modulus, although a clear coorelation between modulus and porosity
was obtained. The mechanical behavior was correlated with the observed microstructure, and fractographic analysis was employed
to elucidate the underlying damage mechanisms.
Solder joint integrity is recognized as a key issue in the reliability of flip chip and ball grid arrays in integrated circuit packages. Significant reductions in the solder joint interconnect size results in both the increased volume fraction of brittle intermetallics in the joint and joule heating due to high current density. Previously, the prevalent failure mode in a solder joint was the ductile thermo-mechanical fracture of solder material due to repeated thermal cycling. In addition to this mode of failure the joints were also found to fail by brittle fracture near the solder–intermetallic interface. To predict and reduce the failure of solder joints, a model that incorporates both plastic damage in bulk solder and solder–intermetallic interface failure is timely and useful to the electronics industry. Based on cohesive fracture theory, a 3D finite element model has been developed to predict the interfacial damage of Sn–Ag–Cu eutectic lead-free solder interconnects. Unified creep-plasticity theory is incorporated in the model considering the creep and hysteresis effects in solder bulk. Using the unified creep-plasticity-cohesive finite element model, the intermetallic layer growth effect has been researched for different solders. The failure of solder is a complex, coupled thermo-electric-mechanical problem. To understand better the phenomenon, a thermo-electric finite element analysis has been conducted to predict the electrical concentration and joule heating effects on the failure of Sn–2.5Ag–0.8Cu–0.5Sb solder under different applied current densities. The temperature and current density distribution in a solder joint with a crack that propagates near the interface of the bulk solder and intermetallic layer has been predicted. Pronounced temperature and electrical current concentration are observed near the crack tip. Although the Pb-free solder is usually regarded as having a higher melting temperature compared with traditional Pb–Sn eutectic solder, the concentration of heat at the crack tip caused by joule heating may still melt the solder material under high current density stressing.