Eutectic Sn-Ag Alloy Research Articles

Chromium improvements on the mechanical performance of a rapidly solidified eutectic Sn–Ag alloy

The eutectic Sn–Ag alloy considered a good replacement to the conventional solders. This study aims to enhance the mechanical performance of the eutectic Sn–Ag alloy in terms of both elasticity and plasticity through investigating the effects of chromium (Cr) content with x = 1, 3, and 5 wt.% on the mechanical and thermal properties of melt-spun Sn–3.5%Ag alloy. Ternary melt-spun Sn–Ag–Cr alloys investigated using x-ray diffractions (XRD), Scanning electron microscope (SEM), Instron machine, Vickers hardness tester, and Differential scanning calorimetry (DSC). The results revealed that the Cr content modifies the microstructure by refining the microstructure of the matrix of Sn solder; suppressing coarse intermetallic compound (IMC) formation; and changing the morphology of the IMC in Sn solder. The high concentration of multi-oriental Ag3Sn that formed with 3 wt.% of Cr, in addition to the compatibility among β-Sn (Tetragonal), Ag3Sn (004) (orthorhombic), and Cr (200) (bcc), enhance the elasticity of melt-spun Sn93.5–Ag3.5–Cr3 solder alloy by 330% (87.98 GPa) and obstructed the fracture process. The tensile test for that melt-spun alloy (3 wt.% of Cr) showed obvious improvements in ultimate tensile strength (UST) by 19% (23.42 MPa), and in toughness by 15.31% (911.15 J m−3) compared to that of the eutectic melt-spun Sn–Ag alloy. Its hardness (Vickers hardness) value has improved by 22.37% (171.50 GPa) compared with that of the eutectic melt-spun Sn–Ag alloy. Slight improvements in thermal performance for that alloy have appeared through thermal conductivity by 8% (49.971 w m−1 k−1) and thermal diffusivity by 85.4% (2.758 × 10–7 m2 s–1) compared to that of the eutectic melt-spun Sn–Ag alloy. All of the above-mentioned improvements are related to the microstructural changes with 3 wt.% of Cr to the eutectic Sn–Ag alloy. The high elasticity, high plasticity, low ∆T, and relatively low melting temperature (Tm) that have been taking place with 3 wt.% of Cr are suggested to increase the mechanical reliability of this alloy to become more desirable in soldering applications and electronic assembly.

Influence of Pd(P) thickness on the Pd-free solder reaction between eutectic Sn-Ag alloy and Au/Pd(P)/Ni(P)/Cu multilayer

Autocatalytic plating deposition of a Pd(P) film over a Ni(P) metallization pad, such as Au/Pd(P)/Ni(P) (electroless nickel/electroless palladium/immersion gold, ENEPIG), has been widely used in microelectronic industry for the past decade. However, the data regarding its soderability with eutectic Sn-Ag (Sn-3.5Ag) is still lacking in the literature. We investigated the effects of Pd(P) thickness (δPd(P)) on the interfacial reaction between Sn-3.5Ag and ENEPIG and the mechanical reliability of the joint system. A slight increase in δPd(P) from 0 μm to 0.2 μm caused the nucleation of a pronounced amount of (Pd,Ni)Sn4 in the solder matrix after reflow, which gradually migrated to the joint interface and formed a (Pd,Ni)Sn4/Ni3Sn4 dual layer after subsequent solid-state aging. The formation of the (Pd,Ni)Sn4 dense layer slowed the growth of Ni3Sn4, facilitated the depletion of the Ni(P) film, and seriously deteriorated the mechanical characteristics of the solder joints. This (Pd,Ni)Sn4-induced embrittlement in solder joints (termed the “Pd-embrittlement phenomenon”) had never been found previously in the Sn-Ag reaction system. The formation and migration of (Pd,Ni)Sn4 was dictated by thermodynamics and could be rationalized using a Pd-Ni-Sn isotherm. The above observations indicated that δPd(P) was an important factor of the ENEPIG solderability and should be strictly controlled to be 0.1 μm or less to prevent the undesired Pd-embrittlement phenomenon in Sn-Ag micro joints.

Chromium effects on the microstructural, mechanical and thermal properties of a rapidly solidified eutectic Sn-Ag alloy

PurposeThis study aims to investigate the chromium (Cr) effects on the microstructural, mechanical and thermal properties of melt-spun Sn-3.5Ag alloy.Design/methodology/approachTernary melt-spun Sn-Ag-Cr alloys were investigated using X-ray diffractions, scanning electron microscope, dynamic resonance technique, instron machine, Vickers hardness tester and differential scanning calorimetry.FindingsThe results revealed that the Ag3Sn intermetallic compound (IMC) and ß-Sn have been refined because of the hard inclusions’ (Cr atoms) effects, causing lattice distortion increasing these alloys. The tensile results of Sn96.4-Ag3.5-Cr0.1 alloy showed an improvement in Young’s modulus more than 100 per cent (42.16 GPa), ultimate tensile strength (UTS) by 9.4 per cent (23.9 MPa), compared with the eutectic Sn-Ag alloy due to the high concentration of Ag3Sn and their uniform distribution. Shortage in the internal friction (Q−1) of about 54 per cent (45.1) and increase in Vickers hardness of about 7.4 per cent (142.1 MPa) were also noted. Hexagonal Ag3Sn formation led to low toughness values compared to the eutectic Sn-Ag alloy, which may have resulted from the mismatching among hexagonal Ag3Sn phase with orthorhombic Ag3Sn and ß-Sn phases. Mechanically, the values of Young’s modulus have been increased, with increasing chromium content, whereas the UTS and toughness values have been decreased. The opposite of this trend appeared in Sn95.8-Ag3.5-Cr0.7 alloy, which may have been due to high lattice distortion (ƹ = 16.5 × 10−4) compared to the other alloys. Increase in the melting temperature Tm, ΔH, Cp and ΔT was because of Ag3Sn IMC formation. The low toughness of Sn96-Ag3.5-Cr0.5 and Sn95.8-Ag3.5-Cr0.7 (109.56 J/m3 and 35.66 J/m3), relatively high melting temperature Tm (223.22°C and 222.65°C) and low thermal conductivity and thermal diffusivity (32.651 w.m−1.k−1 and 0.314 m2/s) make them undesirable in the soldering process. The high UTS, high E, high thermal conductivity and diffusivity, low creep rate and low electrical resistivity, which have occurred with “0.1 Wt.%” of Cr, make this alloy desirable and reliable for soldering applications and electronic assembly.Originality/valueThis study provides chromium effects on the structure of the eutectic Sn-Ag rapidly solidified by melt-spinning technique. In this paper, the authors compared the elastic modulus of the melt-spun compositions, which have been resulted from the Static method with that have been resulted from the Dynamic method. This paper presents new improvements in mechanical and thermal performance.

Copper effects in mechanical, thermal and electrical properties of rapidly solidified eutectic Sn–Ag alloy

The Sn–3.5 wt%Ag alloy considered as a good alternative to Pb–Sn alloys. This study aims to investigate the effects of Cu or Sb additions by 3 or 5 wt% to melt-spun Sn–3.5%Ag alloy. Ternary melt-spun Sn–Ag–Cu and Sn–Ag–Sb alloys investigated using X-ray diffractions (XRD), Scanning electron microscope (SEM), Dynamic resonance technique (DRT), Instron machine, Vickers hardness tester and Differential scanning calorimetry (DSC). The results revealed that the microstructures of the β-Sn phase, Ag3Sn and Cu3Sn intermetallic compounds (IMCs) in the solder matrices were refined due to the effect of Cu additions and melt-spun process. Moreover, increasing Cu content promotes Ag3Sn intermetallic compound (IMC) formation. Consequently, the addition of “3 wt%” of Cu reduced the creep rate ɛ from (3.79 × 10−3) to (1.65 × 10−3) and delayed the fracture point. The tensile results showed an improvement in Young’s modulus by 47% (30.3 GPa), ultimate tensile strength (UST) by 11.6% (23.9 MPa), and in toughness by 20.5% (952.32 J/m3) compared to the eutectic Sn–Ag alloy. Vickers hardness has improved by 3.3% (136.71 MPa) and thermal activation energy by 54% (90.40 KJ/mol) when compared with that of eutectic Sn–Ag alloy. Those improvements are related to the lack of lattice strain from 7.56 × 10−4 without “3 wt%” of Cu to 5.26 × 10−4 with “3 wt%” of Cu. Its melting temperature (Tm) increased by 3 °C due to Ag3Sn IMC increased and Cu3Sn formation, but the pasty rang (mushy zone) decreased by 4 °C with “3 wt%” of Cu. The small lattice strains resulted with “3 wt%” of Cu made the electrical resistivity of this alloy more stable at elevated temperatures. The mechanical, thermal and electrical improvements of Sn93.5–Ag3.5–Cu3 alloy provide good physical performance for soldering process and electronic assembly.

Multiscale modeling of the anisotropic transient creep response of heterogeneous single crystal SnAgCu solder

The lack of statistical homogeneity in functional SnAgCu (SAC) solder joints due to their coarse grained microstructure, in conjunction with the severe anisotropy exhibited by single crystal Sn, renders each joint unique in terms of mechanical behavior. A mechanistic multi-scale modeling framework is proposed in this study to predict the influence of composition and microstructure on the anisotropic transient creep response of single crystal SnAgCu (SAC) solder. Tier I consist of single-crystal eutectic Sn–Ag alloy, with nanoscale Ag3Sn particles embedded in a single-crystal Sn matrix. Tier II consists of single crystal SAC solder which is composed of Sn dendrites surrounded by the eutectic Sn–Ag phase of Tier I. The Tier I anisotropic transient creep model is based on dislocation mechanics. The Tier II model uses the results of Tier I as an input and is based on anisotropic composite micro-mechanics.In Tier I, creep deformation is governed by dislocation impediment and recovery at nanoscale Ag3Sn particles, with recovery being the rate controlling mechanism. Dislocation climb and dislocation detachment at the Ag3Sn particles are proposed to be the competing rate controlling recovery mechanisms. Line tension and mobility of dislocations in dominant slip systems of single crystal Sn are estimated based on the elastic crystal anisotropy of body centered tetragonal (BCT) Sn. The anisotropic transient creep rate of the eutectic Sn–Ag phase of Tier I is then modeled using above inputs and the evolving dislocation density calculated for dominant glide systems during the transient stage of creep. The dominant slip systems are determined based on the dislocation mobility and on the orientation angle between the crystal principal axes and the loading direction. The creep response of the eutectic phase (from Tier 1) is combined with the creep response of Sn lobes at Tier 2, using the anisotropic Mori-Tanaka homogenization theory, to obtain the transient creep response of a SAC305 single crystal along global specimen directions. This model has been calibrated using experimentally obtained transient creep response of a SAC305 single crystal specimen. The above multiscale calibrated model is then used to predict (i) the transient creep response of another SAC305 single crystal specimen and (ii) the effect of orientation (by changing one of the Euler angles) on the transient creep response of SAC305 single crystal. The grain orientation of above two SAC single crystal specimens (with respect to loading direction) were identified with orientation image mapping and then utilized in the model to estimate the resolved shear stress along the dominant slip directions. Parametric studies have also been conducted to predict the effects of the volume fraction, aspect ratio, and orientation of ellipsoidal Sn inclusions on the anisotropic transient creep response of SAC single crystals.

Theoretical and experimental determination of Cu diffusivity in eutectic Sn–Ag system at 235–280 °C

This study evaluated the Cu diffusion and dissolution behaviors in eutectic Sn–Ag alloy with a mathematical model in combination with experimental data. The diffusion was performed at temperatures ranging from 235°C to 280°C with a Cu/Sn–3.5Ag solder bump joint, where the Cu substrate was a solid but the Sn–3.5Ag solder bump was a liquid at all of the temperatures examined. The Cu diffusion coefficient (DCu) with respect to temperature, the activation energy of the Cu diffusion (Q), and the average Cu concentration (Cavg) as a function of diffusion time (t) were investigated via this metallurgical system. The coefficients DCu were 1.2(±0.4)×10−9m2/s (235°C), 2.4(±0.6)×10−9m2/s (250°C), 2.6(±0.7)×10−9m2/s (265°C), and 3.9(±1.6)×10−9m2/s (280°C), which are of the same order as those acquired from the capillary-reservoir technique for the bulk Cu/Sn system by Ma and Swalin in 1969. The activation energy Q was estimated based on the Arrhenius relation and was approximately 13.3kJ/mol at 235–280°C, reflecting the insensitivity of DCu to temperature. A calculated Cavg–t profile predicted that Cavg approaches saturation after 70s (235°C), which agrees well with experiments. The modeling results were consistent with the experimental and literature data, indicating that the proposed mathematical model is an effective method to represent the Cu diffusion/dissolution behavior in a confined solder alloy.

Formation of SnAg solder bump by multilayer electroplating

A novel method, multilayer electroplating, was proposed to prepare alloy bumps. It consists to electroplating different structural elements of alloys sequentially and then forming uniform alloy through reflowing. The formation of eutectic SnAg alloy bumps was taken as an example to verify the applicability of the method in this study. Near-eutectic SnAg solder bumps were formed successfully through multilayer electroplating. After reflowing for 1min at 250^oC, SnAg alloy with homogeneous structure and composition was obtained from multilayer structure. Fine Ag3Sn particles with uniform distribution were precipitated inside the solder and Cu6Sn5 intermetallic compounds (IMCs) in scallop-like shape formed at the SnAg solder/Cu interface. As reflow time increasing, the morphology of Ag3Sn remained unchanged but the thickness of Cu6Sn5 layer increased. The characteristic of multilayer electroplated SnAg solder bumps was similar to that of traditional alloy, which indicated that it was a suitable method to form SnAg alloy solder bumps.

Electrodeposition and Characterization of Eutectic Sn–Ag Alloy as Solder Bumps for Flip-Chip Interconnection

Eutectic Sn 3.5 wt % Ag alloy is one of the most promising lead-free solders in wafer bumping for flip-chip interconnection. Its electrodeposition in pyrophosphate-iodide-based baths was investigated. In particular, the effects of polyethylene glycol 600 and formaldehyde as surfactants were studied, and their addition proved to facilitate achieving a near-eutectic composition for the Sn–Ag alloy. Different deposited film morphologies were analyzed using scanning electron microscopy (SEM) and focused ion beam/SEM. X-ray diffraction investigations indicated that a biphasic structure of and was present in the as-electrodeposited film. To further confirm the efficacy of the electrodeposition process, fine pitch Sn–Ag solder bumps with smooth surface and homogeneous composition were achieved on glass test wafers.

Reliability of Solder Joint with Sn–Ag–Cu–Ni–Ge Lead-Free Alloy under Heat Exposure Conditions

The reliability of a solder ball joint with a Sn-Ag-Cu-Ni-Ge lead-free alloy, which is expected to be an advanced lead-free solder, was investigated under heat exposure conditions. Solder ball joints with a eutectic Sn-Ag alloy and a ternary Sn-Ag-Cu alloy were also prepared to compare with that of the Sn-Ag-Cu-Ni-Ge alloy. Microstructual observations of the cross sections of the solder ball joints were conducted to investigate microstructural evolutions in the solders and the growth kinetics of reaction layers formed at joint interfaces. The influence of heat exposure treatment on joint strength was investigated by ball shear test. Moreover, the influence of surface treatment of a Cu pad on the reliability of the solder joint was also investigated.

The constitutive creep equation for a eutectic Sn-Ag alloy using the modified theta-projection concept

Creep data for a eutectic tin-silver alloy at temperatures between 298 K and 398 K have been analyzed using the modified theta-projection concept, instead of the steady-state creep constitutive equation in the following formula: ɛcr=A {1−exp(−αt)}+B {exp(αt)−1}, where A, B, and α are constants to be experimentally determined. The equation describes well the creep curves of the eutectic tin-silver alloy up to the tertiary stage. All constants exhibited power law relationships with the applied stress. The rate constant, α, has a high stress exponent, which is attributed to dispersion strengthening. The rate constant a and the strain factor B only showed temperature dependence, while the strain factor A was independent of temperature. The activation energy for α was 65 kJ/mol at high stresses and 90 kJ/mol at low stresses. The energies suggest that the dislocation pipe diffusion and the lattice diffusion are predominant at high stresses and low stresses, respectively.

Improvements of microstructure, wettability, tensile and creep strength of eutectic Sn–Ag alloy by doping with rare-earth elements

To improve the properties of the eutectic Sn–Ag lead-free solder alloy, various amounts of mixed rare-earth (RE) elements, mainly Ce and La, were added. The microstructure, wetting properties, melting behavior, mechanical properties, and creep behavior were studied. It was revealed that RE elements can refine the intermetallics, and with 0.5% RE addition, the RE-bearing phase can be detected in the microstructure of the slow-cooled alloy. The results of differential scanning calorimetry indicate that the melting points of the RE-doped alloys are slightly lower than that of the Sn–3.5Ag and have a eutectic peak. The wetting property and creep resistance of the Sn–3.5Ag–0.25RE alloy are better than those of the Sn–3.5Ag alloy. The creep properties were studied at the temperatures of 303, 348, and 393 K, at various stress levels between 8 and 34 MPa. The stress exponents of the Sn–3.5Ag and Sn–3.5Ag–0.25RE were obtained at these temperatures. Tensile, creep, and wetting properties were found to improve with the addition of RE elements. The improvement of creep resistance is due to the fine dispersion of intermetallics and the decrease in interface energy between matrix and intermetallics. The wettability improvement is mainly due to the accumulation of RE elements at the solder/flux interface, leading to the reduction of the interfacial tension between solder and flux.

２‐Ｓｔｅｐめっき法によるＳｎ‐Ａｇはんだバンプ形成プロセス

The eutectic Sn-3.5wt.%Ag solder alloy is expected to be one of the candidates for Pb-free solder bump materials. But it is difficult for the conventional Sn-Ag alloy plating method to obtain eutectic composition with good uniformity within wafer, and to control the composition of plating solution for long term product run. Therefore, we have developed the 2-step electro-plating bump process using separate plating reactors for Ag and Sn. The eutectic Sn-Ag alloy bumps were easily obtained by annealing the Ag/Sn metal stacks sequentially electroplated. It is easy to control the Sn-Ag alloy composition by changing the thickness ratio of Ag layer to Sn layer in an Ag/Sn metal stack. Furthermore, the 2-step electroplating bump process does not need to control the content ratio of Ag to Sn in a plating solution. The eutectic Sn-Ag solder bump process described here is confirmed to be easily applied to mass production.

Eutectic Sn-Ag solder bump process for ULSI flip chip technology

A novel eutectic Pb-free solder bump process, which provides several advantages over conventional solder bump process schemes, has been developed. A thick plating mask can be fabricated for steep wall bumps using a nega-type resist with a thickness of more than 50 /spl mu/m by single-step spin coating. This improves productivity for mass production. The two-step electroplating is performed using two separate plating reactors for Ag and Sn. The Sn layer is electroplated on the Ag layer. Eutectic Sn-Ag alloy bumps can be easily obtained by annealing the Ag/Sn metal stack. This electroplating process does not need strict control of the Ag to Sn content ratio in alloy plating solutions. The uniformity of the reflowed bump height within a 6-in wafer was less than 10%. The Ag composition range within a 6-in wafer was less than /spl plusmn/0.3 wt.% Ag at the eutectic Sn-Ag alloy, analyzed by ICP spectrometry. SEM observations of the Cu/barrier layer/Sn-Ag solder interface and shear strength measurements of the solder bumps were performed after 5 times reflow at 260/spl deg/C in N/sub 2/ ambient. For the Ti(100 nm)/Ni(300 nm)/Pd(50 nm) barrier layer, the shear strength decreased to 70% due to the formation of Sn-Cu intermetallic compounds. Thicker Ti in the barrier metal stack improved the shear strength. The thermal stability of the Cu/barrier layer/Sn-Ag solder metal stack was examined using Auger electron spectrometry analysis. After annealing at 150/spl deg/C for 1000 h in N/sub 2/ ambient, Sn did not diffuse into the Cu layer for Ti(500 nm)/Ni(300 nm)/Pd(50 nm) and Nb(360 nm)/Ti(100 nm)/Ni(300 nm)/Pd(50 nm) barrier metal stacks. These results suggest that the Ti/Ni/Pd barrier metal stack available to Sn-Pb solder bumps and Au bumps on Al pads is viable for Sn-Ag solder bumps on Cu pads in upcoming ULSIs.

Effect of thermal ageing on (Sn-Ag, Sn-Ag- Zn)/PtAg, Cu/Al2O3 solder joints

As-fabricated solders of eutectic Sn-Ag and ternary Sn-3.5 wt% Ag-1 wt% Zn alloys are coupled with metallized substrates including PtAg/Al2O3 and Cu/Al2O3 to simulate the solder joint in microelectronics. The growth mechanism of intermetallics and the mechanical properties of solder joints after thermal ageing (150 °C and 200 °C) are evaluated. In this study, a 1206 passive device/solder/metallization/Al2O3 surface mount technology (SMT) assembly is employed, and a Cu stud is attached on the ceramic substrate assembly to evaluate mechanical properties and the fracture morphology by the pull-off test. In addition, microstructure evolution of the interfacial morphology, elemental and phase distribution are probed with the aid of scanning electron microscopy (SEM), electron probe micro-analysis (EPMA) and X-ray diffraction (XRD) techniques. There are two intermetallics (Cu3Sn and Cu6Sn5) formed at the eutectic Sn-Ag solder/Cu metallized layer interface, while only Cu6Sn5 is observed in the Sn-3.5 Ag-1Zn/Cu system. However, in the PtAg metallized substrate, only Ag3Sn is present, regardless of which solders are employed. Cu6Sn5 and Ag3Sn in the Sn-3.5 Ag-1Zn system contain 2–5 at% Zn due to the higher solubility of Zn in both Cu and Ag. The adhesion strength decreases as the time increases for all solder joint systems in the thermal ageing test. The solder joint with eutectic Sn-Ag alloy exhibits higher fracture load than that with Sn-3.5 Ag-1Zn alloy. From the fracture surface analysis, as the ageing time increases, the fracture takes place from the solderconductor interface toward the inside of the IMC (intermetallic compound). © 1998 Kluwer Academic Publishers