I. Dutta

Creep behavior of interfaces in fiber reinforced metal–matrix composites

Abstract The elevated temperature deformation behavior of interfaces in model single fiber composites was isolated and studied using a fiber push-down approach, whereby the interface is loaded in shear. Two fiber–matrix systems, one with no mutual solubility (quartz–lead) and the other with limited mutual solubility (nickel–lead), were investigated. In both systems, the matrix and fiber underwent sliding relative to each other, with the interface acting as a high diffusivity path. The mechanism of sliding was inferred to be interface-diffusion-controlled diffusional creep with a threshold stress (Bingham flow). The behavior was modeled analytically using a continuum approach, and an expression for the constitutive creep behavior of the interface was derived. The model provided a physical basis for the observed threshold behavior, which was found to be directly related to the normal (radial) residual stress acting on the fiber–matrix interface. The results are deemed to be significant because: (1) in some instances, interfacial sliding may be instrumental in determining the overall creep/thermal cycling response of a composite; and (2) they offer an alternative rationalization of threshold behavior during diffusional flow (besides interface reaction control) and may be useful in understanding creep in multi-phase systems with internal stresses.

Role of interfacial and matrix creep during thermal cycling of continuous fiber reinforced metal-matrix composites

Abstract A uni-dimensional micro-mechanical model for thermal cycling of continuous fiber reinforced metal–matrix composites is developed. The model treats the fiber and matrix as thermo-elastic and thermo-elasto-plastic-creeping solids, respectively, and allows the operation of multiple matrix creep mechanisms at various stages of deformation through the use of unified creep laws. It also incorporates the effect of interfacial sliding by an interface-diffusion-controlled diffusional creep mechanism proposed earlier (Funn and Dutta, Acta mater. , 1999, 47 , 149). The results of thermal cycling simulations based on a graphite fiber reinforced pure aluminum–matrix composite were compared with experimental data on a P100 graphite–6061 Al composite. The model successfully captured all the important features of the observed strain responses of the composite for different experimental conditions, such as the observed heating/cooling rate dependence, strain hysteresis, residual permanent strain at the end of a cycle, as well as both intrusion and protrusion of the fiber-ends relative to the matrix at the completion of cycling. The analysis showed that the dominant deformation mechanism operative in the matrix changes continually during thermal cycling due to continuous stress and temperature revision. Based on these results, a framework for the construction of a transient deformation mechanism map for thermal excursions of continuous fiber composites is proposed.

Thermal cycling studies of a cross-plied P100 graphite fibre-reinforced 6061 aluminium composite laminate

Response to thermal cycling of a 0/90 cross-plied P100 Gr-6061 aluminium composite laminate was studied between a minimum temperature (Tmin) of 25 ‡C and maximum temperatures (Tmax) of 100 and 540 ‡C. Strain hysteresis was observed between the heating and cooling half-cycles and was attributed to anelastic strains induced by matrix residual stresses. A residual plastic strain was also observed after the first cycle, and was seen to disappear after subsequent cycles. Alteration of the thermal residual stress state of the matrix via heat treatments was found to change significantly the magnitude of the plastic strain. These results were compared with those of studies on unidirectionally reinforced P100 Gr-6061 aluminium composites, and the differences were explained on the basis of the residual stresses resident in the matrix. Optical and electron microscopy were also utilized to observe thermal damage, which occurred predominantly along improperly bonded fibre-matrix interfaces.

Kinetic evidence for the structural similarity between a supercooled liquid and an icosahedral phase in Zr65Al7.5Ni10Cu12.5Ag5 bulk metallic glass

By differential scanning calorimetric measurement, the kinetics of the phase transformation present in Zr65Al7.5Ni10Cu12.5Ag5 bulk metallic glass during continuous heating was investigated. It was found that the effective activation energy from a supercooled liquid to an icosahedral quasicrystalline phase is much lower than that from the supercooled liquid to eutectic crystalline phases. In addition, the activation energy from the icosahedral phase to the crystalline phases is almost the same as that from the supercooled liquid to the crystalline phases. Both of them support that the local atomic structure is similar for the supercooled liquid and the icosahedral phase in the bulk metallic glass.

Miniaturized impression creep testing of ball grid array solder balls attached to microelectronic packaging substrates

This article reports on the design and implementation of a miniaturized impression creep apparatus for characterizing the creep behavior of tiny solder balls attached to a ball grid array (BGA) microelectronic packaging substrate. The technique requires no special sample preparation, can probe individual solder balls, and proffers high data throughput by allowing numerous creep curves to be obtained from one substrate, as well as by minimizing the time required to achieve steady state creep. The apparatus reported here uses a 100-μm-diameter cylindrical WC punch to characterize the creep behavior of 750-μm-diameter BGA solder balls from ambient temperature to 423 K. A video imaging system facilitates precise alignment and placement of the indenter on the specimen at the test temperature. The possible effect of substrate curvature on the experimental solder creep curves was evaluated and was deemed to be insignificant. Example creep curves and data based on 90Pb-10Sn BGA solder balls are presented. The tes...

Fracture of Sn-Ag-Cu Solder Joints on Cu Substrates:I. Effects of Loading and Processing Conditions

During service, microcracks form inside solder joints, making microelectronic packages highly prone to failure on dropping. Hence, the fracture behavior of solder joints under drop conditions at high strain rates and under mixed-mode conditions is a critically important design consideration for robust joints. This study reports on the effects of joint processing and loading conditions on the microstructure and fracture response of Sn-3.8%Ag-0.7%Cu (SAC387) solder joints attached to Cu substrates. The impact of parameters which control the microstructure (reflow condition, aging) as well as loading conditions (strain rate and loading angle) are explicitly studied. A methodology based on the calculation of the critical energy release rate, GC, using compact mixed-mode (CMM) samples was developed to quantify the fracture toughness of the joints under conditions of adhesive (i.e., interface-related) fracture. In general, higher strain rate and increased mode-mixity resulted in decreased GC. GC also decreased with increasing dwell time at reflow temperature, which produced a thicker intermetallic layer at the solder–substrate interface. Softer solders, produced by slower cooling following reflow, or post-reflow aging, showed enhanced GC. The sensitivity of the fracture toughness to all of the aforementioned parameters reduced with an increase in the mode-mixity. Fracture mechanisms, elucidating the effects of the loading conditions and process parameters, are briefly highlighted.

Microstructurally Adaptive Model for Primary and Secondary Creep of Sn-Ag-Based Solders

Sn-Ag-based solders are susceptible to appreciable microstructural coarsening due to the combined effect of thermal and mechanical stimuli during service and storage. This results in evolution of the creep properties of the solder over time, necessitating the development of a thermo-mechanical history-dependent creep model for accurate prediction of the long-term reliability of microelectronic solder joints. In this paper, the coarsening behavior of Ag3Sn and Cu6Sn5 precipitates in ball grid array-sized joints of Sn-3.8Ag-0.7Cu solder attached to Ni bond-pads with four different thermo-mechanical histories is reported. Because of the substantial numerical superiority of Ag3Sn over Cu6Sn5, it was inferred that the evolution of mechanical properties during aging is controlled largely by the coarsening of Ag3Sn. An effective diffusion length (x̅) for Ag diffusion in Sn was defined, and it is shown to adequately describe the thermo-mechanical history dependence of Ag3Sn particle size. The shear creep behavior of these joints was experimentally characterized, and the entire creep data were fitted to a unified model combining exponential primary creep and power-law steady state creep. The parameter x̅ was then incorporated into the creep equation to produce a unified creep model, which can adapt to thermo-mechanical history-dependent microstructural coarsening in the solder. Predictions using this creep law show very good agreement with experimental creep data for several different test and microstructural conditions.

Impression creep testing and microstructurally adaptive creep modeling of lead free solder interconnects

Creep plays an important role in the reliability of solder joints under thermo-mechanical fatigue conditions encountered by a microelectronic package during service. In addition, the fine intermetallic precipitates (Ag/sub 3/Sn and/or Cu/sub 6/Sn/sub 5/) in the microstructures of the new lead-free solders (Sn-Ag and Sn-Ag-Cu) can undergo significant in situ strain-enhanced coarsening during TMC, resulting in in-service evolution of the creep behavior of the joints. Since there are significant microstructural/compositional differences between bulk solder samples and tiny microelectronic solder joints, it is critical to develop accurate creep testing methodologies on tiny life-sized solder joints and microstructurally adaptive constitutive creep models for the emerging Pb-free solder alloys. In this paper, we present creep data obtained from tests conducted on individual Sn/sub 4/Ag/sub 0.5/Cu ball grid array (BGA) solder balls attached to a packaging substrate, using a newly developed miniaturized impression creep apparatus, which affords high test throughput with minimal sample preparation. Coarsening of intermetallic particles is demonstrated to influence creep behavior in two ways. At low stresses, the creep rate increases proportionately with precipitate size. At high stresses, precipitate coarsening influences creep response by altering the threshold stress for particle-limited creep. Based on the experimental observations, a microstructurally adaptive creep model, which accounts for the effects of coarsening on the creep response of solder joints, and is capable of adjusting itself as solder joint microstructures evolve during service, is presented, along with experimental determination of the relevant coarsening kinetics parameters.

Atomic force microscopy study of plastic deformation and interfacial sliding in Al thin film: Si substrate systems due to thermal cycling

A method is proposed to measure the plastic deformation of thin metallic films on Si substrates induced by thermal cycling. The cross-sectional profiles of pattern-grown square Al films with a thickness of ∼250 nm and a size of ∼6 μm×6 μm were measured before and after thermal cycling by employing an atomic force microscope. With the assistance of statistical analysis, the change in the size and shape of the thin films were determined. Based on theoretical considerations, the thermal cycling deformation of thin films is attributed to creep and plasticity effects, accommodated by diffusion-controlled interfacial sliding.

Diffusionally accommodated interfacial sliding in metal-silicon systems

The kinetics and mechanism of diffusionally accommodated interfacial sliding (interfacial creep) under far-field shear and normal stresses was studied, based on diffusion-bonded Al-Si-Al sandwich specimens. A previously developed interfacial creep law [Funn and Dutta, Acta Mater 1999; 47: 149], which proposed that interfaces may slide via interface-diffusion controlled diffusional creep, was experimentally validated by carrying out a systematic parametric study. In agreement with the model, the Si-Al interfaces slid via diffusional creep (n = 1) under the influence of an effective shear stress, which depends on the far-field shear and normal stresses, as well as the interfacial topography. Compressive stresses acting normal to the interface lowered the effective shear stress, resulting in a threshold effect, thus reducing the sliding rate. The rate of sliding was controlled by diffusional mass transport through a thin amorphous, O-rich interfacial layer, under the influence of local interfacial stress gradients, which arose due to the topological features of the interface. Instances of interfacial sliding in the absence of interfacial de-cohesion, which have been noted in composites, thin-film systems, etc., may be explained by the present mechanism, which also offers an alternative rationalization of threshold behavior during diffusional flow (besides interface-reaction control).