Modelling the behavior of solder joints for wafer level SiP

Design for manufacture of system-in-package (SiP) structures is dependent on a number of physical processes that affect the final quality of the package in terms of its performance and reliability. Solder joints are key structures in a SiP and their behavior can be the critical factor in terms of reliability. This paper discusses the results from a research programme on design for manufacturing of system in package (SiP) technologies. The focus of the paper is on thermo-mechanical modelling of solder joints. This includes the behavior of the joints during testing plus some important insights into the reflow process and how physical phenomena taking place at the assembly stage can affect solder joint behavior. Finite element analysis of a numerical model of an SiP structure with various design parameters is discussed. The goal of this analysis is to identify the most promising combination of design parameters which guarantee longer lifetime of the solder joints and hence the SiP component. The parameters that were studied are the size of the package (i.e. number of solder joints per row), the presence of the underfill and/or the reinforcement as well as the thickness of the passive die. Discussion was also provided on phenomena that take place during the reflow process where the solder joints are formed. In particular, the formation of intermetallics at the solder-pad interfaces

Effects of thermal cycling profiles on the performance of chip-on-flex assembly using anisotropic conductive films

Anisotropic conductive films (ACFs) are widely used in the electronic packaging industries because of their fine pitch potential and the assembly process is simpler compared to the soldering process. However, there are still unsolved issues in the volume productions using ACFs. The main reason is that the effects of many factors on the interconnects are not well understood. This work focuses on the performance of ACF-bonded chip-on-flex assemblies subjected to a range of thermal cycling test conditions. Both experimental and three-dimensional finite element computer modelling methods are used. It has been revealed that greater temperature ranges and longer dwell-times give rise to higher stresses in the ACF interconnects. Higher stresses are concentrated along the edges of the chip-ACF interfaces. In the experiments, the results show that higher temperature ranges and prolonged dwell times increase contact resistance values. Close examination of the microstructures along the bond-line through the scanning electron microscope (SEM) indicates that cyclic thermal loads disjoint the conductive particles from the bump of the chip and/or pad of the substrate and this is thought to be related to the increase of the contact resistance value and the failure of the ACF joints.

High electric current density-induced interfacial reactions in micro ball grid array (μBGA) solder joints

The effect of a high electric current density on the interfacial reactions of micro ball grid array solder joints was studied at room temperature and at 150 °C. Four types of phenomena were reported. Along with electromigration-induced interfacial intermetallic compound (IMC) formation, dissolution at the Cu under bump metallization (UBM)/bond pad was also noticed. With a detailed investigation, it was found that the narrow and thin metallization at the component side produced “Joule heating” due to its higher resistance, which in turn was responsible for the rapid dissolution of the Cu UBM/bond pad near to the Cu trace. During an “electromigration test” of a solder joint, the heat generation due to Joule heating and the heat dissipation from the package should be considered carefully. When the heat dissipation fails to compete with the Joule heating, the solder joint melts and molten solder accelerates the interfacial reactions in the solder joint. The presence of a liquid phase was demonstrated from microstructural evidence of solder joints after different current stressing (ranging from 0.3 to 2 A) as well as an in situ observation. Electromigration-induced liquid state diffusion of Cu was found to be responsible for the higher growth rate of the IMC on the anode side.

Effect of 0.5 wt % Cu addition in Sn–3.5% Ag solder on the dissolution rate of Cu metallization

The dissolution of thin film under-bump-metallization (UBM) by molten solder has been one of the most serious processing problems in electronic packaging technology. Due to a higher melting temperature and a greater Sn content, a molten lead-free solder such as eutectic SnAg has a faster dissolution rate of thin film UBM than the eutectic SnPb. The work presented in this paper focuses on the role of 0.5 wt % Cu in the base Sn–3.5%Ag solder to reduce the dissolution of the Cu bond pad in ball grid array applications. We found that after 0.5 wt % Cu addition, the rate of dissolution of Cu in the molten Sn–3.5%Ag solder slows down dramatically. Systematic experimental work was carried out to understand the dissolution behavior of Cu by the molten Sn–3.5%Ag and Sn–3.5%Ag–0.5%Cu solders at 230–250 °C, for different time periods ranging from 1 to 10 min. From the curves of consumed Cu thickness, it was concluded that 0.5 wt % Cu addition actually reduces the concentration gradient at the Cu metallization/molten solder interface which reduces the driving force of dissolution. During the dissolution, excess Cu was found to precipitate out due to heterogeneous nucleation and growth of Cu6Sn5 at the solder melt/oxide interface. In turn, more Cu can be dissolved again. This process continues with time and leads to more dissolution of Cu from the bond pad than the amount expected from the solubility limit, but it occurs at a slower rate for the molten Sn–3.5%Ag–0.5%Cu solder. © 2003 American Institute of Physics.

Reliability study of the electroless Ni–P layer against solder alloy

Copper (Cu) has been widely used in the under bump metallurgy of chip and substrate metallization for chip packaging. However, due to the rapid formation of Cu–Sn intermetallic compound (IMC) at the tin-based solder/Cu interface during solder reaction, the reliability of this type of solder joint is a serious concern. In this work, electroless nickel–phosphorous (Ni–P) layer was deposited on the Cu pad of the flexible substrate as a diffusion barrier between Cu and the solder materials. The deposition was carried out in a commercial acidic sodium hypophosphite bath at 85 °C for different pH values. It was found that for the same deposition time period, higher pH bath composition (mild acidic) yields thicker Ni–P layer with lower phosphorous content. Solder balls having composition 62%Sn–36%Pb–2%Ag were reflowed at 240 °C for 1 to 180 min on three types of electroless Ni–P layers deposited at the pH value of 4, 4.8 and 6, respectively. Thermal stability of the electroless Ni–P barrier layer against the Sn–36%Pb–2%Ag solder reflowed for different time periods was examined by scanning electron microscopy equipped with energy dispersed X-ray. Solder ball shear test was performed in order to find out the relationship between the mechanical strength of solder joints and the characteristics of the electroless Ni–P layer deposited. The layer deposited in the pH 4 acidic bath showed the weak barrier against reflow soldering whereas layer deposited in pH 6 acidic bath showed better barrier against reflow soldering. Mechanical strength of the joints were deteriorated quickly in the layer deposited at pH 4 acidic bath, which was found to be thin and has a high phosphorous content. From the cross-sectional studies and fracture surface analyses, it was found that the appearance of the dark crystalline phosphorous-rich Ni layer weakened the interface and hence lower solder ball shear strength. Ni–Sn IMC formed at the interfaces was found to be more stable at the low phosphorous content (∼14 at.%) layer. Electroless Ni–P deposited at mild acidic bath resulting phosphorous content of around 14 at.% is suggested as the best barrier layer for Sn–36%Pb–2%Ag solder.

Multiple reflow study of ball grid array (BGA) solder joints on Au/Ni metallization

This paper evaluates the shearing behavior of ball grid array (BGA) solder joints on Au/Ni/Cu pads of FR4 substrates after multiple reflow soldering. A new Pb-free solder, Sn–3Ag–0.5Cu–8In (SACI), has been compared with Sn–3Ag–0.5Cu (SAC) and Sn–37Pb (SP) solders, in terms of fracture surfaces, shearing forces and microstructures. Three failure modes, ball cut, a combination of solder shear and solder/pad bond separation, and pad lift, are assessed for the different solders and reflow cycles. It is found that the shearing forces of the SP and SAC solder joints tend to increase slightly with an increase in the number of reflow cycles due to diffusion-induced solid solution strengthening of the bulk solder and augmentation of the shearing area. However, the shearing forces of the SACI solder joints decrease slightly after four cycles of reflow, which is ascribed to the thermal degradation of both the solder/intermetallic compound (IMC) and IMC/Ni interfaces. The SACI solder joints yield the highest strengths, whereas the SP solder joints give the smallest values, irrespective of the number of reflow cycles. Thickening of the interfacial IMC layer and coarsening of the dispersing IMC particles within the bulk solders were also observed. Nevertheless, the variation of shearing forces and IMC thickness with different numbers of reflow cycles was not so significant since the Ni under layer acted as an effective diffusion barrier. In addition, the initially-formed IMC layer retarded the further extensive dissolution of the pad material and its interaction with the solder

Shear strength analysis of ball grid array (BGA) solder interfaces

Ball shear test is the most common test method used to assess the reliability of bond strength for ball grid array (BGA) packages. In this work, a combined experimental and numerical study was carried out to realize of BGA solder interface strength. Solder mask defined bond pads on the BGA substrate were used for BGA ball bonding. Different bond pad metallizations and solder alloys were used. Solid state aging at 150degC up to 1000 h has been carried out to change the interfacial microstructure. Cross-sectional studies of the solder-to-bond pad interfaces was conducted by scanning electron microscopy (SEM) equipped with an energy dispersive X-ray (EDX) analyzer to investigate the interfacial reaction phenomena. Ball shear tests have been carried out to obtain the mechanical strength of the solder joints and to correlate shear behaviour with the interfacial reaction products. An attempt has been taken to realize experimental findings by Finite Element Analysis (FEA). It was found that intermetallic compound (IMC) formation at the solder interface plays an important role in the BGA solder bond strength. By changing the morphology and the microchemistry of IMCs, the fracture propagation path could be changed and hence, reliability could be improved

Effect of current stressing on the reliability of 63Sn37Pb solder joints

The effect of current stressing on the reliability of 63Sn37Pb solder joints with Cu pads was investigated at temperatures of −5 °C and 125 °C up to 600 h. The samples were stressed with 3 A current (6.0 × 102 A/cm2 in the solder joint with diameter of 800 μm and 1.7 × 104 A/cm2 in the Cu trace with cross section area of 35 × 500 μm). The temperatures of the samples and interfacial reaction within the solder joints were examined. The microstructural change of the solder joints aged at 125 °C without current flow was also evaluated for comparison. It was confirmed that the current flow could cause the temperature of solder joints to rise rapidly and remarkably due to accumulation of massive Joule heat generated by the Cu trace. The solder joints stressed at 125 °C with 3 A current had an extensive growth of Cu6Sn5 and Cu3Sn intermetallic compounds (IMC) at both top and bottom solder-to-pad interfaces. It was a direct result of accelerated aging rather than an electromigration or thermomigration effect in this experiment. The kinetic is believed to be bulk diffusion controlled solid-state reaction, irrespective of the electron flow direction. When stressed at −5 °C with 3 A current, no significant change in microstructure and composition of the solder joints had occurred due to a very low diffusivity of the atoms as most Joule heat was eliminated at low temperature. The IMC evolution of the solder joints aged at 125 °C exhibited a subparabolic growth behavior, which is presumed to be a combined mechanism of grain boundary diffusion and bulk diffusion. This is mainly ascribed to the retardant effect against the diffusion course by the sufficiently thick IMC layer that was initially formed during the reflow soldering.

Failure mechanisms of ACF joints under isothermal ageing

The possible failure mechanisms of anisotropic conductive film (ACF) joints under isothermal ageing conditions have been identified through experiments. It has been found that ACF joints formed at higher bonding temperatures can prevent increases in the contact resistance for any ageing temperature. The higher the ageing temperature the higher the electrical failure rate is. The formation of conduction gaps between the conductive particles and the pads and damages to the metal coatings of the particle have been identified as the reasons behind the electrical failures during ageing. In order to understand the mechanism for the formation of the conduction gap and damages in metal coatings during the isothermal ageing, computer modelling has been carried out and the results are discussed extensively. The computer analysis shows that stresses concentrate at the edges of the particle–pad interface, where the adhesive matrix meets the particle. This could lead to subsequent damages and reductions in the adhesion strength in that region and it is possible for the conductive particle to be detached from the pad and the adhesive matrix. It is believed that because of this a conduction gap appears. Furthermore, under thermal loading the thermal expansion of the adhesive matrix squeezes the conductive particle and damages the metal coatings. Experimental evidences support this computational finding. It is, therefore, postulated that if an ACF-based electronic component operates in a high temperature aging condition, its electrical and mechanical functionalities will be at risk.

Investigation of wall-slip effect on paste release characteristic in flip-chip stencil printing process

As the trend toward further miniaturisation of pocket and handheld consumer electronic products continues apace, the requirements for even smaller solder joints will continue. With further reductions in the size of solder joints, the reliability of solder joints will become more and more critical to the long-term performance of electronic products. Solder joints play an important role in electronics packaging, serving both as electrical interconnections between the components and the board, and as mechanical support for components. With world-wide legislation for the removal/reduction of lead and other hazardous materials from electrical and electronic products, the electronics manufacturing industry has been faced with an urgent search for new lead-free solder alloy systems and other solder alternatives. In order to achieve high volume, low cost production, the stencil printing process and subsequent wafer bumping of solder paste has become indispensable. There is wide agreement in industry that the paste printing process accounts for the majority of assembly defects, and most defects originate from poor understanding of the effect of printing process parameters on printing performance. The printing of ICAs and lead-free solder pastes through the very small stencil apertures required for flip chip applications was expected to result in increased stencil clogging and incomplete transfer of paste to the printed circuit pads. Paste release from the stencil apertures is dependent on the interaction between the solder paste, surface pad and aperture wall; including its shape. At these very narrow aperture sizes the paste rheology becomes crucial for consistent paste withdrawal because for smaller paste volumes surface tension effects become dominant over viscous flow. Successful aperture filling and release will greatly depend on the rheology of the paste material. Wall-slip plays an important role in characterising the flow behaviour of solder paste materials. The wall- slip arises due to the various attractive and repulsive forces acting between the solder particles and the walls of the measuring geometry. These interactions could lead to the presence of a thin solvent layer adjacent to the wall, which gives rise to slippage. The wall slip effect can play an important role in ensuring successful paste release after the printing process. The aim of this study was to investigate the influence of the paste microstructure on slip formation for the paste materials (lead-free solder paste and isotropic conductive adhesives). The effect of surface roughness on the paste viscosity was investigated. It was also found that altering the surface roughness of the parallel plate measuring geometry did not significantly eliminate wall slip as was expected. But results indicate that the use of a relatively rough surface helps to increase paste adhesion to the plates, inducing structural breakdown of the paste. Most importantly, the study also demonstrated on how the wall slip formation in the paste material could be utilised for understanding of the paste microstructure and its flow behaviour

Effect of long-term aging on the rheological characteristics and printing performance of lead-free solder pastes used for flip-chip assembly

Stencil printing of solder pastes is a critical stage in the SMT assembly process as a high proportion of the solder-related defects can be attributed to this stage. As the trend towards product miniaturization continues, there is a greater need for better understanding of the rheological behaviour and printing performance of new paste formulations. This fundamental understanding is crucial for achieving the repeatable solder paste deposits from board-to-board and pad-to-pad required for more reliable solder interconnections. The paper concerns a study on the effect of ageing on the rheological characteristics and printing performance of new lead-free solder pastes formulations used for flip-chip assembly applications. The objective is to correlate the rheological characteristics of aged paste samples to their printing performance. The methodology developed can be used for bench-marking new lead-free paste formulations in terms of shelf life, the potential deterioration in rheological characteristics and their printing performance.

The Effects of Cilostazol on Exercise-induced Ischaemia-reperfusion Injury in Patients with Peripheral Arterial Disease

Objectives: Cilostazol improves walking distance in peripheral arterial disease (PAD) patients. The study objectives were to assess the effects of cilostazol on walking distance, followed by the additional assessment of cilostazol on exercise-induced ischaemiaereperfusion injury in such patients.

Methods: PAD patients were prospectively recruited to a double-blinded, placebo-controlled trial. Patients were randomised to receive either cilostazol 100 mg or placebo twice a day. The primary end-point was an improvement in walking distance. Secondary end-points included the assessment of oxygen-derived free-radical generation, antioxidant consumption and other markers of the in?ammatory cascade. Initial and absolute claudication distances (ICDs and ACDs, respectively) were measured on a treadmill. In?ammatory response was assessed before and 30 min post-exercise by measuring lipid hydroperoxide, ascorbate, atocopherol, b-carotene, P-selectin, intracellular and vascular cell-adhesion molecules (I-CAM and V-CAM), thromboxane B2 (TXB2), interleukin-6, interleukin-10, high-sensitive C-reactive protein (hsCRP), albuminecreatinine ratio (ACR) and urinary levels of p75TNF receptor. All tests were performed at baseline and 6 and 24 weeks.

Results: One hundred and six PAD patients (of whom 73 were males) were recruited and successfully randomised from December 2004 to January 2006. Patients who received cilostazol demonstrated a more signi?cant improvement in the mean percentage change from baseline in ACD (77.2% vs. 26.6% at 6 weeks, pZ0.026 and 161.7% vs. 79.0% at 24 weeks, pZ0.048) as compared to the placebo. Cilostazol reduced lipid hydroperoxide levels compared to a placebo-related increase before and after exercise (6 weeks: pre-exercise: 11.8% vs.

Accelerating Multiple Sequence Alignment with the Cell BE Processor

The Cell Broadband Engine (BE) Architecture is a new heterogeneous multi-core architecture targeted at compute-intensive workloads. The architecture of the Cell BE has several features that are unique in high-performance general-purpose processors, most notably the extensive support for vectorization, scratch pad memories and explicit programming of direct memory accesses (DMAs) and mailbox communication. While these features strongly increase programming complexity, it is generally claimed that significant speedups can be obtained by using Cell BE processors. This paper presents our experiences with using the Cell BE architecture to accelerate Clustal W, a bio-informatics program for multiple sequence alignment. We report on how we apply the unique features of the Cell BE to Clustal W and how important each is in obtaining high performance. By making extensive use of vectorization and by parallelizing the application across all cores, we demonstrate a speedup of 24.4 times when using 16 synergistic processor units on a QS21 Cell Blade compared to single-thread execution on the power processing unit. As the Cell BE exploits a large number of slim cores, our highly optimized implementation is just 3.8 times faster than a 3-thread version running on an Intel Core2 Duo, as the latter processor exploits a small number of fat cores.

Assessment of the impedance cardiogram recorded by an automated external defibrillator during clinical cardiac arrest

OBJECTIVE: To assess the impedance cardiogram recorded by an automated external defibrillator during cardiac arrest to facilitate emergency care by lay persons. Lay persons are poor at emergency pulse checks (sensitivity 84%, specificity 36%); guidelines recommend they should not be performed. The impedance cardiogram (dZ/dt) is used to indicate stroke volume. Can an impedance cardiogram algorithm in a defibrillator determine rapidly circulatory arrest and facilitate prompt initiation of external cardiac massage?

DESIGN: Clinical study.

SETTING: University hospital.

PATIENTS: Phase 1 patients attended for myocardial perfusion imaging. Phase 2 patients were recruited during cardiac arrest. This group included nonarrest controls.

INTERVENTIONS: The impedance cardiogram was recorded through defibrillator/electrocardiographic pads oriented in the standard cardiac arrest position.

MEASUREMENTS AND MAIN RESULTS: Phase 1: Stroke volumes from gated myocardial perfusion imaging scans were correlated with parameters from the impedance cardiogram system (dZ/dt(max) and the peak amplitude of the Fast Fourier Transform of dZ/dt between 1.5 Hz and 4.5 Hz). Multivariate analysis was performed to fit stroke volumes from gated myocardial perfusion imaging scans with linear and quadratic terms for dZ/dt(max) and the Fast Fourier Transform to identify significant parameters for incorporation into a cardiac arrest diagnostic algorithm. The square of the peak amplitude of the Fast Fourier Transform of dZ/dt was the best predictor of reduction in stroke volumes from gated myocardial perfusion imaging scans (range = 33-85 mL; p = .016). Having established that the two pad impedance cardiogram system could detect differences in stroke volumes from gated myocardial perfusion imaging scans, we assessed its performance in diagnosing cardiac arrest. Phase 2: The impedance cardiogram was recorded in 132 "cardiac arrest" patients (53 training, 79 validation) and 97 controls (47 training, 50 validation): the diagnostic algorithm indicated cardiac arrest with sensitivities and specificities (+/- exact 95% confidence intervals) of 89.1% (85.4-92.1) and 99.6% (99.4-99.7; training) and 81.1% (77.6-84.3) and 97% (96.7-97.4; validation).

CONCLUSIONS: The impedance cardiogram algorithm is a significant marker of circulatory collapse. Automated defibrillators with an integrated impedance cardiogram could improve emergency care by lay persons, enabling rapid and appropriate initiation of external cardiac massage.

Lorentz TEM investigation of the switching behavior in artificial spin ice building blocks

Frustration – the inability to simultaneously satisfy all interactions – occurs in a wide range of systems including neural networks, water ice and magnetic systems. An example of the latter is the so called spin-ice in pyrochlore materials [1] which have attracted a lot of interest not least due to the emergence of magnetic monopole defects when the ‘ice rules’ governing the local ordering breaks down [2]. However it is not possible to directly measure the frustrated property – the direction of the magnetic moments – in such spin ice systems with current experimental techniques. This problem can be solved by instead studying artificial spin-ice systems where the molecular magnetic moments are replaced by nanoscale ferromagnetic islands [3-8]. Two different arrangements of the ferromagnetic islands have been shown to exhibit spin ice behaviour: a square lattice maintaining four moments at each vertex [3,8] and the Kagome lattice which has only three moments per vertex but equivalent interactions between them [4-7]. Magnetic monopole defects have been observed in both types of lattices [7-8]. One of the challenges when studying these artificial spin-ice systems is that it is difficult to arrive at the fully demagnetised ground-state [6-8].
Here we present a study of the switching behaviour of building blocks of the Kagome lattice influenced by the termination of the lattice. Ferromagnetic islands of nominal size 1000 nm by 100 nm were fabricated in five island blocks using electron-beam lithography and lift-off techniques of evaporated 18 nm Permalloy (Ni80Fe20) films. Each block consists of a central island with four arms terminated by a different number and placement of ‘injection pads’, see Figure 1. The islands are single domain and magnetised along their long axis. The structures were grown on a 50 nm thick electron transparent silicon nitride membrane to allow TEM observation, which was back-coated with a 5 nm film of Au to prevent charge build-up during the TEM experiments.
To study the switching behaviour the sample was subjected to a magnetic field strong enough to magnetise all the blocks in one direction, see Figure 1. Each block obeys the Kagome lattice ‘ice-rules’ of “2-in, 1-out” or “1-in, 2-out” in this fully magnetised state. Fresnel mode Lorentz TEM images of the sample were then recorded as a magnetic field of increasing magnitude was applied in the opposite direction. While the Fresnel mode is normally used to image magnetic domain structures [9] for these types of samples it is possible to deduce the direction of the magnetisation from the Lorentz contrast [5]. All images were recorded at the same over-focus judged to give good Lorentz contrast.
The magnetisation was found to switch at different magnitudes of the applied field for nominally identical blocks. However, trends could still be identified: all the blocks with any injection pads, regardless of placement and number, switched the direction of the magnetisation of their central island at significantly smaller magnitudes of the applied magnetic field than the blocks without injection pads. It can therefore be concluded that the addition of an injection pad lowers the energy barrier to switching the connected island, acting as a nucleation site for monopole defects. In these five island blocks the defects immediately propagate through to the other side, but in a larger lattice the monopoles could potentially become trapped at a vertex and observed [10].
References

[1] M J Harris et al, Phys Rev Lett 79 (1997) p.2554.
[2] C Castelnovo, R Moessner and S L Sondhi, Nature 451 (2008) p. 42.
[3] R F Wang et al, Nature 439 (2006) 303.
[4] M Tanaka et al, Phys Rev B 73 (2006) 052411.
[5] Y Qi, T Brintlinger and J Cumings, Phys Rev B 77 (2008) 094418.
[6] E Mengotti et al, Phys Rev B 78 (2008) 144402.
[7] S Ladak et al, Nature Phys 6 (2010) 359.
[8] C Phatak et al, Phys Rev B 83 (2011) 174431.
[9] J N Chapman, J Phys D 17 (1984) 623.
[10] The authors gratefully acknowledge funding from the EPSRC under grant number EP/D063329/1.