Effects of minor Fe, Co, and Ni additions on the reaction between SnAgCu solder and Cu

doi:10.1016/j.jallcom.2008.11.052

Journal of Alloys and Compounds

Volume 478, Issues 1–2, 10 June 2009, Pages 121-127

https://doi.org/10.1016/j.jallcom.2008.11.052 Get rights and content

Abstract

The reactions between Cu and the Sn2.5Ag0.8Cu solders doped with 0.03 wt.% Fe, Co, or Ni were studied. Reaction conditions included multiple reflows for up to 10 times and solid-state aging at 160 °C for up to 2000 h. In the multiple reflow study, Cu₆Sn₅ was the only reaction product noted for all the different solders used. Reflows using the solder without doping produced a thin, dense layer of Cu₆Sn₅. Adding Fe, Co, or Ni transformed this microstructure into a much thicker Cu₆Sn₅ with many small trapped solder regions between the Cu₆Sn₅ grains. In the solid-state aging study, both Cu₆Sn₅ and Cu₃Sn formed, but adding Fe, Co, or Ni produced a much thinner Cu₃Sn layer. Because the Cu₃Sn growth had been linked to the formation of micro voids, which in turn increased the potential for a brittle interfacial fracture, thinner Cu₃Sn layers might translate into better solder joint strength.

Introduction

In the past few years, the SnAgCu family of solders has obtained a wide acceptance as a replacement for the PbSn eutectic solder in electronic applications. At present, relatively fewer activities are undertaken by researchers worldwide to develop another solder family to replace the PbSn eutectic solder. Instead, the current main research thrust in lead-free solder alloy development is on enhancing or fine-tuning the various properties of SnAgCu through adding minor alloying elements. For example, Ni was evaluated for its potential as a minor alloying element to Sn-based lead free solders [1], [2], [3], [4], [5], [6], [7], [8], and so were Ge [2], [5], Fe [9], Co [6], [8], [9], Zn [5], [10], [11], Bi [12], Mn [5], Ti [5], Si [5], and Cr [5]. These previous studies point out minor alloy additions can strengthen the solder performance in many different respects. One of the more noteworthy alloying elements is Ni. It was shown Ni addition to Sn3.5Ag (3.5 wt.% Ag, balance Sn) in amounts as minute as 0.1 wt.% could substantially hinder the Cu₃Sn growth during soldering as well as during the following solid-state aging [1], [7]. The Cu₃Sn growth had been linked to the formation of micro voids, which in turn increased the potential for brittle interfacial fracture [13]. Recently, it was shown that drop test performance increased for solders joints with just a small amount of Ni addition (<1 wt.%) [14].

The objective of this study is to examine whether Fe and Co additions have a similar effect as Ni does. Emphasis is placed on a systematic comparison study on the effect of Fe, Co, and Ni additions. Further, the level of alloy addition is reduced to as low as 0.03 wt.% so the minimum effective addition of each element can be assessed. Solders with simultaneous Fe and Ni addition as well as simultaneous Co and Ni addition are also prepared to examine whether there is any interaction between the alloying elements.

Section snippets

Experimental

Solder balls with six different compositions, Sn2.5Ag0.8Cu, Sn2.5Ag0.8Cu0.03Fe, Sn2.5Ag0.8Cu0.03Co, Sn2.5Ag0.8Cu0.03Ni, Sn2.5Ag0.8Cu0.03Fe0.03Ni, and Sn2.5Ag0.8Cu0.03Co0.03Ni (all compositions in weight percentage), were prepared from 99.999% purity elements. The solder ball compositions were verified using ICP-AES (inductively coupled plasma atomic emission spectroscopy), and it was estimated the reported compositions had a maximum 0.005 wt.% uncertainty. For a given solder ball composition,

Reaction during reflow

Fig. 1 shows the backscattered electron micrographs for the samples with different alloy additions that were reflowed for one time. The Cu₆Sn₅ phase is clearly visible in all cases. At this stage, the Cu₆Sn₅ phase was the only phase noted at the interface, and Cu₃Sn was not observed. In the following, Sn2.5Ag0.8Cu will be denoted as SAC, and Sn2.5Ag0.8Cu0.03Fe will be denoted as SAC0.03Fe, etc. Comparing Fig. 1(a) to Fig. 1(b–f), one notes adding Fe, Co, or Ni to SAC made the amount of Cu₆Sn₅

Reaction during reflow

This study showed doping the SAC with a small amount of Fe, Co, or Ni would not change the type of the reaction product with the Cu substrates. The reaction product was always Cu₆Sn₅ during a typical reflow. The Cu₃Sn phase only formed when the reflow time became very long [1] or during the solid-state aging. This differed from the reaction between Ni and the Cu-doped Sn-based solders, which was very sensitive to the Cu concentration [19], [20], [21], [22]. Under this situation [19], [20], [21]

Summary

(1)
In the study of multiple reflow using the Sn2.5Ag0.8Cu, Sn2.5Ag0.8Cu0.03Fe, Sn2.5Ag0.8Cu0.03Co, Sn2.5Ag0.8Cu0.03Ni, Sn2.5Ag0.8Cu0.03Fe0.03Ni, and Sn2.5Ag0.8Cu0.03Co0.03Ni solder over Cu substrate, Cu₆Sn₅ was the only reaction product for all the different solders used.
(2)
Reflows using the solder without doping produced a thin, dense layer of Cu₆Sn₅. Adding Fe, Co, or Ni transformed the microstructure into a much thicker Cu₆Sn₅ with many small trapped solder regions between the grains.
(3)
The amount of

Acknowledgments

This work was supported by the National Science Council of R.O.C. through grant NSC-95-2221-E-002-443-MY3. The EPMA analysis performed by Ms. S. Y. Tsai of National Tsing Hua University is also acknowledged.

References (24)

F. Gao et al.
Mater. Sci. Eng. A
(2006)
J.Y. Tsai et al.
J. Electron. Mater.
(2003)
C.M. Chung et al.
J. Electron. Mater.
(2003)
H. Nishikawa et al.
J. Electron. Mater.
(2006)
H. Yu et al.
I.E. Anderson et al.
J. Electron. Mater.
(2006)
F. Gao et al.
J. Electron. Mater.
(2006)
C.E. Ho et al.
J. Mater. Sci. Mater. Electron.
(2007)
F. Gao et al.
J. Electron. Mater.
(2008)
I.E. Anderson et al.
J. Electron. Mater.
(2004)

S.K. Kang et al.

J. Electron. Mater.

(2006)

S.C. Yang et al.

J. Mater. Res.

(2006)

Cited by (170)

Equilibrium phase diagram design and structural optimization of SAC/Sn-Pb composite structure solder joint for preferable stress distribution
2023, Materials Characterization
The Package-on-Package (PoP) technology is an efficient knowhow to increase the efficiency of electronic components, hence to secure the continuation of the validity of Moore's law. This novel form of packaging technology with bottom stackable narrow pitch Ball Grid Array (BGA) contributes to the integrity of the package. However, the high solder joint density introduces a great challenge on the packaging reliability. Therefore, more stringent requirements must be put forward for achieving the increased performance. That can be achieved by using composite solder joints. In this study, a novel composite SAC (Sn-3Ag-0.5Cu)/Sn-Pb (Sn37Pb) solder joint was designed for improving the stress distribution and its overall mechanical integrity. The formation process and a phase distribution of composite solder joint were investigated, indicating that the phase distribution mainly depends on the reflow temperature. The dissolution ratio of SAC into the composite solder at different temperatures was estimated by the phase diagram. The predicted values were consistent with the experimental results. A review of mechanical property test results, the best performance solder joint could be obtained at 200 °C for the dwell time of 60 s if the composite includes 10 wt% SnPb. The stress distribution of composite solder joint was analyzed by finite element simulation. By actively concentrating the stress in the middle area of the solder joint, the maximum stress in the composite solder reduced for up to 29% when compared to full SAC solder joint.
Microstructural and mechanical characterization of Cu/Sn SLID bonding utilizing Co as contact metallization layer
2022, Journal of Alloys and Compounds
Most micro-electro-mechanical systems (MEMS) devices contain fragile moving parts, which poses challenges in process integration of interconnection methods requiring wet-chemistry, such as solid-liquid interdiffusion bonding (SLID). These sensitive MEMS structures can be protected from either the wet-chemistry or plated metals during chemical/electro-chemical plating of SLID interconnection materials; however, this is a complex process. Hence, our previous research has investigated employing a physically deposited contact metallization on the wafers containing functional devices instead of chemically deposited layers (such as electrochemical Cu). Co is a plausible contact metallization layer for Cu-Sn SLID bonding, as it is chemically compatible with Cu–Sn systems. Furthermore, it can positively impact the mechanical reliability of the intermetallic compounds (IMCs) due to the stabilizing of the HT-hexagonal Cu₆Sn₅ phase down to room temperature and suppressing the Cu₃Sn phase formation and subsequent void formation. However, it is critical to control Co thickness to achieve a stable bond based on our previous research on Co bulk in contact with Cu-Sn electroplated silicon chips. To utilize Co as a contact metallization layer for wafer-level Cu-Sn SLID bonding, it is necessary to define appropriate metal layers in the contact metallization stack. Consequently, the present study investigated four different contact metallization stacks including A) 40nmTi/100 nm Co, B) 40 nm Ti/200 nm Mo/100 nm Co, C) 40 nm Ti/500 nm Co, and D) 40 nm Ti/200 nm Mo/500 nm Co. More specifically, we evaluated the microstructural formation and evolution and mechanical performance of the joints. Our study revealed that the Ti/Mo/100 nm Co contact metallization stack for (4 µm)Cu/(2 µm)Sn SLID bonding is composed of IMCs (Cu,2.5at%Co)₆Sn₅ and Cu₃Sn without remaining Sn. Moreover, the joint contained a negligible number of voids even after long-time annealing at 150 °C. Our analysis of the mechanical properties of the joint showed that 1) the tensile fracture surface exhibited a mixture of ductile and brittle fractures, and 2) the Young’s modulus of (Cu,2.5at%Co)₆Sn₅ was higher than Cu₆Sn₅, while hardness of (Cu,2.5at%Co)₆Sn₅ and Cu₆Sn₅ were comparable. By employing a Ti/Mo/100 nm Co contact metallization stack, the current study was able to produce 25 µm and 10 µm void free (4 µm Cu/2 µm Sn) microbumps.
Forming mechanism and growth of Kirkendall voids of Sn/Cu joints for electronic packaging: A recent review
2022, Journal of Advanced Joining Processes
For the common Cu-Sn interconnection system in microelectronics packaging, a significant concern is the sporadic interfacial void formation within the Cu–Sn interface and intermetallic compound (IMC) layer during the service. The existence of these voids affects the electrical and mechanical properties of the joint and thus deteriorates reliability. Most scholars simply attribute the Kirkendall voids problem to the Kirkendall effect caused by the difference in Cu/Sn inter-diffusion coefficients. However, Kirkendall voids are formed for a combination of many reasons. The plating additives, current density, surface roughness, coating thickness, and substrate types affect the formation of Kirkendall voids. We believe that among the many factors affecting the formation of voids, the unbalanced diffusion of elements (Kirkendall effect) is the root cause, and impurities and Cu microstructure are the dominant factors. The addition of alloying elements to the solder affects the formation of voids by changing the unbalanced diffusion of elements. This paper releases the possible mechanism of Kirkendall voids and gives the perspective summary from three steps of void formation, and the available suppression method is given based on the final solution.
Experimental investigation and thermodynamic calculation of the Co–Sn–Bi ternary system
2022, Calphad: Computer Coupling of Phase Diagrams and Thermochemistry
Phase relations of the Co–Sn–Bi ternary system have been studied experimentally for the whole composition range at 250 and 450 °C by means of scanning electron microscopy (SEM) coupled with wave dispersive spectrometer (WDS) and x-ray diffraction (XRD). Five three-phase regions have been confirmed in this system at 250 °C, and three three-phase regions exist in 450 °C. No ternary compound can be observed at these two temperatures. Moreover, Bi has almost no solubility in the binary compounds αCo₃Sn₂, CoSn, CoSn₂ and CoSn₃. The primary solidification phases were also distinguished. Based on the experimental results, the thermodynamic calculations for Co–Bi and Co–Sn–Bi system were conducted by using the CALPHAD technique. A set of self-consistent thermodynamic parameters for the Co–Sn–Bi ternary system was obtained, and the experiment results are in good agreement with the calculated results.
Investigation of the microstructural evolution and detachment of Co in contact with Cu–Sn electroplated silicon chips during solid-liquid interdiffusion bonding
2022, Journal of Alloys and Compounds
Solid-liquid interdiffusion (SLID) bonding is one of the most promising novel methods for micro-(opto)-electromechanical system (MEMS/MOEMS) wafer-level packaging. However, the current SLID bonding solutions require the use of an electrochemical deposition method for MEMS/MOEMS wafers as well, which significantly complicates the process integration options. Hence, this work proposes Co as a potential option for compatible contact metallization on MEMS/MOEMS wafers to utilize mature Cu–Sn SLID bonding. The focus of this study is on gaining a fundamental understanding of the microstructural formation and evolution of Co substrates in contact with Cu–Sn electroplated silicon wafers and identifying possible failures of joints during bonding, which are prerequisites for guaranteeing devices’ manufacturability, functionality, and long-term reliability. The effect of bonding time and temperature on the microstructural evolution and phase formation of Co substrates in contact with Cu–Sn electroplated silicon chips was investigated. Moreover, a phase diagram of the Co–Cu–Sn ternary system was thermodynamically evaluated based on experimental data. Samples were successfully bonded at 250 °C for 1500 and 2000 s and at 280 °C for 1000 s. The main interfacial intermetallic compounds were identified as (Cu,Co)₆Sn₅, Cu₃Sn, and (Co,Cu)Sn₃. Co stabilized the high-temperature hexagonal (η) Cu₆Sn₅ phase down to room temperature. Bond detachment was observed when applying either a higher bonding temperature or a longer bonding time. Two critical factors that cause detachment during bonding were recognized: first, a change in thermodynamic equilibrium when exceeding the maximum allowed Co content in Cu₆Sn₅ formed adjacent to the CoSn₃ phase and a discontinuous change in the Co content in the Cu6Sn5 grown on the Cu and Co sides; second, stress exerted due to the rapid growth of (Co,Cu)Sn₃ between the Co substrate and (Cu,Co)₆Sn₅. Therefore, achieving successful bonding in the Co–Sn–Cu SLID system requires governing the amount of dissolved Co atoms in liquid Sn and the CoSn₃ formation, both of which can be achieved by manipulating the relative thickness of the Co, Cu, and Sn layers. The observations and calculations in this work show that a prerequisite for obtaining successful bonding in the Co–Sn–Cu SLID system at 250 °C is a Co–to–Sn thickness ratio of 0.04 or less.
Improving tensile strength of SnAgCu/Cu solder joint through multi-elements alloying
2021, Materials Today Communications
Citation Excerpt :
In contrast, although there is a little difference in thickness, the interfacial IMC layers in the multi-alloyed solder/Cu joints are thicker but show a loose porous structure (see Fig. 8(b)-(e)), and the Ni and Sb elements can be detected in the IMC layers, because Ni can replace some Cu atoms in the Cu6Sn5 and Sb can dissolve into the interfacial IMCs [3,10,19]. Formation of a porous interfacial IMC structure should be related with the addition of Ni, and similar phenomenon has been reported at the interfaces between Cu and Ni-containing solders by early literatures [20,21]. The rough IMC/solder interface will increase the mechanical bonding between the solder and the IMC layers, which will further affect the fracture behavior of the solder joint.
In this study, different content combinations of the Sb, Bi, Ni, In and Ce elements were added into the SnAgCu (SAC) solder, then the composite alloying effects of these elements on microstructure and properties of the SAC solder, and the interfacial reaction and tensile strength of the SAC/Cu solder joints were investigated. The results reveal that the melting ranges of the multi-alloyed solders are larger, and their wetting areas increase by different degree. More intermetallic compounds (IMCs) and some Bi particles are formed in the multi-alloyed solders and the solder grains are refined, which significantly increase the microhardness. The interfacial IMC layer change from single layered scallop-type Cu₆Sn₅ grains at the SAC/Cu interface to an agglomeration of fine (Cu, Ni)₆Sn₅ particles by the multi-element alloying, while the residual stress in the interfacial IMC layers can be released due to the decrease in the IMC grain size. Through solid solution strengthening, precipitation strengthening and fine grain strengthening of the solder and release of residual stress in the interfacial IMCs, tensile strength of the multi-alloyed SAC/Cu solder joints was significantly improved to 100–160 MPa, and the fracture location transfer from inside the solders close to the joint interface to inside the interfacial IMCs layer.

View all citing articles on Scopus

View full text

Effects of minor Fe, Co, and Ni additions on the reaction between SnAgCu solder and Cu

Abstract

Introduction

Section snippets

Experimental

Reaction during reflow

Reaction during reflow

Summary

Acknowledgments

Mater. Sci. Eng. A

J. Electron. Mater.

J. Electron. Mater.

J. Electron. Mater.

J. Electron. Mater.

J. Electron. Mater.

J. Mater. Sci. Mater. Electron.

J. Electron. Mater.

J. Electron. Mater.

J. Electron. Mater.

J. Mater. Res.