Enhancing Cu-Cu Stitch Bonding Reliability on Cu Leadframe Through Ultrathin Cu-Selective Passivation

Despite advances in 2.5D and 3D packaging, wire bonding remains the industry’s most widely used interconnect method due to its cost-effectiveness. Ultrasonic wedge bonding, also known as stitch bonding, is a critical interconnection technique that links IC dies to external circuits for power and signal transmission, where precise control of bonding parameters is essential to ensure reliable connectivity. However, as the industry transitions from traditional gold (Au) wire to cost-effective copper (Cu) and palladium-coated copper (PCC) wire, new challenges have emerged in ensuring second bond (stitch) reliability, particularly on modern leadframe substrates with silver (Ag) coatings. While Ag plating helps suppress Cu oxidation, it adds to material cost and raises environmental concerns due to its bioaccumulation potential. To address this, we developed a low-cost, ultrathin Cu-selective passivation coating that effectively suppresses Cu oxidation and enables direct Cu-Cu stitch bonding without the need for Ag. This passivation layer is fully compatible with standard IC packaging processes and does not interfere with surrounding dielectrics or interconnects. This paper discusses the application of this ultrathin coating on copper-based leadframe (LF) and is subjected to stitch bonding with 25 μm (1 mil) aluminum (Al-1wt%Si) and palladium-coated copper wire (PCC) under different heating conditions (170°C-200°C). The bonding power, time, and force are kept constant throughout the work. Coated Cu leadframes demonstrated a substantial improvement in pull strength and reliability. Notably, bonds on bare-Cu LF failed at the interface due to oxidation, while passivated-Cu LF exhibited failure at the wire neck, indicating stronger bond integrity. Preliminary data with PCC wire at 125°C further support these findings. This work highlights a promising, environmentally sustainable alternative to Ag plating, enabling reliable Cu-Cu stitch bonding for next-generation, low-cost IC packaging solutions.

Fine pitch copper wire bonding presents challenges to both first and second bond processes. Corrosion of the first bond copper-aluminum (Cu-Al) intermetallic compound bond interface layer can be induced by mobile chlorine in the epoxy mold compound, and the second bond process window can be narrower than with gold wire. Palladium-coated copper wire is believed to overcome these two problems, however, the actual benefits and challenges of the wire need to be considered by the semiconductor industry. The price of palladium-coated copper wire is 2.5–3 times higher than bare copper wire. The mechanical properties of palladium-coated copper wire increase the risk of damaging bond pad structures if not bonded correctly. The addition of a thin palladium layer can increase electrical resistivity, which can be a concern for high frequencies and smaller diameter wire applications. Others promote palladium-coated copper wire with reports of improved biased highly accelerated stress test (HAST) results versus bare copper wire. As a consequence, debate over the choice between palladium-coated and bare copper wire is common. Some semiconductor suppliers and original equipment manufacturers (OEMs) incorrectly believe that palladium-coated copper wire is a panacea for all historical concerns with the use of bare copper wire. A study has been conducted to assess advantages and disadvantages of bare and palladium-coated copper. This paper shows bare copper wire can provide the same level of chlorine-induced corrosion resistance as palladium-coated copper wire if the copper-aluminum intermetallic bond is properly formed and the mold compound is correctly formulated. The basis for the belief that palladium-coated copper wire provides better resistance to chlorine-induced corrosion is explained and the electrical performance difference between palladium-coated and bare copper wires is discussed. High temperature (175°C) storage life testing, up to 7000 hours, was conducted with both wire types to determine the end-of-life failure mechanism. Bond interface cracking, initiating in the bond periphery, was observed. The time dependence of copper-aluminum intermetallic phase transformation for both wire types will be presented. It is shown that choice criteria for each wire type can be defined by product field application requirements and not by perceived advantages. The reported work shows bare copper has equal performance to palladium-coated copper wire under Automotive Electronic Council (AEC) reliability grade 1 in specified package types when the bonding process, substrate / lead frame design and mold compound have been correctly optimized [1].

Palladium coated copper (PCC) wire bonding on roughen preplated leadframe is a preferred electrical interconnect in the semiconductor package due to its robustness package performances. PCC wire is replacing bare copper wires to eliminate the copper oxidation problem. Whereas, roughen treated preplated leadframe is used to eliminate post-plating of leadframe as well as to enhance the adhesion of leadframe to the epoxy mold compounds. However, role of the plating thickness of roughen preplated leadframe on the PCC wire has not yet been fully understood. Thus, this paper investigates stitch bonding strength of PCC wire on the AuAg/Pd/Ni/Cu preplated leadframe with different AuAg plating thickness. Stitch pull test results on PCC wire bonding on different preplated leadframe showed higher stitch bonding strength for samples with higher AuAg plating thickness. Optical micrographs showed mechanical failure at bond heel of test sample, due to higher tensile stress during the pull test. Thicker AuAg layers may provide greater cushioning effect against wire deformation at bond heel during stitch bonding process. This in turns increased the stitch bonding strength of the PCC wire bond. STEM micrograph and EDX line-scan profile in all samples showed formation of interdiffusion zone at the PCC wire/ leadframe bonding interface. At the distinct bonding interface, interdiffusion zone extended from Pd layer of PCC wire to Pd layer of leadframe, where Pd species from both sides were intermixed. Whereas at blurred bonding interface, interdiffusion zone was extended further to Pd/Ni interface region. Formation of interdiffusion zone may strengthened wire bond between the PCC wire and the leadframe, thus prevented bond delamination during the stitch pull strength test.

There is a growing interest in copper wire bonding for IC interconnection due to the cost savings and better material properties. Recently, palladium-coated copper (Pd-Cu) wire has emerged as the preferred choice over bare Cu wire in fine wire diameter applications, especially with BGA packages. The application of Pd-Cu wire is a solution to prevent copper oxidation during the bonding process. This is true in the Second Bond but may not be the situation when it comes to First Bond. This is because during the Free Air Ball (FAB) formation, the Pd-Cu wire is re-melted and the Pd distribution on the FAB depends on various factors. It has been reported that Pd distribution on the FAB is very much dependent on the effect of EFO (Electronic Flame Off) current and also the wire type. To ensure best protection of First Bond against HAST and PCT tests, it is best to have the Pd distributed over the entire surface of the FAB, thereby forming a protective armour against corrosion attack by halogen ion present in mold compound chemistry. Different types of Pd-Cu wires will need different FAB formation optimization. If the EFO parameter is not optimized properly, it will result in dimple FAB or/and inconsistent FAB that is not perfectly round-shaped ball, with a deep valley at the centre that shape like “blueberry”. In this paper, two different types of Pd-Cu wires and its FAB formation at different EFO conditions were investigated. Various analysis techniques, Optical Scope, SEM, EDX and FIB were used to analyze the Pd distributions in the Pd-Cu FAB. A general guideline of how to obtain consistent good round FAB for Pd-Cu wire is presented. A method of analyzing the good FAB with Pd layer protection has been demonstrated. This methodology will help to ensure a good and proper Pd-Cu wire FAB to start with before carrying out First Bond wire bond optimization.

Copper (Cu) wire bonding technology had been widely accepted as a interconnect material in semiconductor packaging. The main advantage of Cu as a interconnect material is cost, thermal and electrical performance as compare to gold. However effect of corrosive elements on bare Cu wire remain a big challenge for the reliability of Cu wire packages. Market trend is pushing for Palladium coated copper (PCC) wire to be the alternative for bare Cu wire, due to the better bondability process and resistance to corrosive elements. Nevertheless there are many researches suggest PCC wire is not suitable for High Temperature Storage (HTS) reliability stress test more than 150 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">o</sup> C due to the intrinsic degradation of PCC wire. From literature review the main contribution to the degration of PCC wire was due to the crack or void on the Palladium (Pd) coating. Cu underneath the Pd coating will diffuse through the crack to the surface and weaken the integrity of the wire. Through sparking parameter optimization, optimal Pd coverage on the FAB can be achieved hence first bond integrity can be secured. On the other hand, 100% Pd coverage on the second bond wedge is impossible to obtain due to the extensive mechanical contact of capillary on the wedge during formation. In this paper, the study will focus on the wedge PCC wire on roughen leadframe with NiPdAuAg surface finishing. The study will cover difference 3 main factors influencing the integrity of wedge, i) wedge formation, ii) leadframe AuAg plating thickness iii) ionic elements from mold compound. Sample with various wedge formation on different AuAg plating thickness with and without mold compound will be subjected to HTS 200 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">o</sup> C up to 900 hours and comparison of the degration rate for different samples can be identify. The aim of the study is to identify the dominant factor of the PCC wedge degration on HTS stress so that improvement for package beyond AEC Q100 Grade 0 can be achieved.

Microstructural characterization of alloyed palladium coated copper wire under high temperature

Simulation study on the effect of palladium layer thickness and temperature on the bonding properties of palladium coated copper wire

Chip to chip Cu wire bonding studies have been carried out on TSLP package, with bare Cu and palladium coated copper (PCC) wire. This paper discusses on feasibility studies of Cu wire chip to chip bonding which include bump cut mode, scale lead setting and Cu bump oxidation risk assessment. From the results, it is shown that bump cut mode has significant effect to bonding results. Appropriate bump cut mode has to be selected for PCC wire in order to have a palladium coating on the bump cut surface. The PCC wire stitch pull is higher than bare Cu wire at both 0 hour and HTS168hours (200°C). Nevertheless, scale lead setting for stitch bond positioning is crucial to obtain optimum stitch bond length on bump, also avoid bonding at bump edge and wire short die. Evaluation on bump oxidation was also carried out to study the effect of Cu bump oxidation to stitch-bump interface. It is observed that higher Oxygen content on Cu bump after heat staging more than 5s without N <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> gas protection. Void line is observed between stitch-bump interface for both 5s and 60s heat staging sample. No void line observed for PCC wire up to 60s heat staging. Void line between stitch-bump interface is further studied with HTS168hrs (200°C) and HTS500hrs (175°C) for void growth. Not degradation of stitch pull readings and void growth is observed.

In this study, we experimented with palladium-coated copper (PdCu) wires of different diameters (0.6 mil, 0.7 mil and 0.8 mil) and from different manufactures. It was seen that wire diameters that varied within the manufacturing tolerances show an effect on bonding responses. The effect was especially prominent in the production environment, for example, when electric-flame-off (EFO) time remained fixed and free air ball (FAB) and subsequent bonded ball size varied according to the variation in wire diameter. The effect of Pd thickness control in PdCu wire was also studied and its implication to bonding responses is presented. Aside from the variations from the wires, bonding parameters such as EFO conditions, cover gas types, and gas flow rates were also found to affect wire bonding responses by varying degrees for wires from different suppliers. All of these observations suggest the added complications and considerations necessary when dealing with PdCu wire. Therefore, ensuring an understanding of the wire characteristics, especially pertaining to the Pd coating and their relations to wire bonding responses, is critical in overcoming these additional challenges.

Alternative solutions to gold wire are being sought as a for lower cost packages. Palladium coated copper (PCC) wires serve as a potential alternative to increase package robustness and lower package cost particularly in the automotive industry. Palladium on the surface of the wire promotes better adhesion of the wire to the surface where it is bonded. In effect, this may result to elimination of broken wire at heel and lifted ball. This paper discusses the material considerations involved in shifting to PCC wire. In this paper, the benefits of coating bare Cu wire with Pd are discussed. These improvements include enhancements in material properties, wirebond responses, intermetallic compound (IMC) formation and reliability that Pd coating imparts on Cu wire. Palladium coating on Cu wire results to better first and second bond reliability hence, improving package robustness.

Palladium coated copper wire has been introduced in the semiconductor wirebond industry to addressed issues related to bare copper wire, mainly fast oxidation. The presence of the palladium coat provides protection to oxidation of bare copper wire core for a short period of time. This is very useful especially on bond stitch on ball (BSOB) wirebonding where a stitch bond is placed on top of bump ball. The relative hardness of the coated copper wire however possess significant challenge for large scale BSOB wirebonding. Conventional gold (4N Au) wirebonding is the best choice but with rising cost the alternative version is preferred by the industry. Silver (88, 92, 95%Ag) alloy wires offers cheaper cost solution but still remain to be proven for high reliability wirebonding. Alloying also of Ag wires results to an increase in the resistivity of the wire. This is unwanted especially on Mosfet devices where RDSon resistance is significantly controlled. Combining cost and reliability performance requirements, palladium coated copper wire is the still best choice.

Effects of alloying elements in high reliability copper wire bond material for high temperature applications

Since around 2008, the shift from Gold (Au) bonding wire to Copper (Cu) bonding wire has been taking place, full scale, with the aim of reducing costs. When compared with Au, Cu wire presents challenges in reliability and repeatable bonding characteristics in terms of chemical stability, which is required in high reliability applications. Therefore Cu wire adoption in automotive and industrial semiconductors has been limited. Conventionally the market for Cu bonding wires has been divided into two types: bare Cu wires (high purity) and Palladium coated copper (PCC) bonding wires. These wires have yet to satisfy the required characteristics for high reliability products such as industrial and automotive electronics. A new breed of alternative bonding wires has been developed to offer performance advantages for high reliability applications compared to bare copper wire and PCC wire. Cu alloy wire and Ag alloy wires continue their market introduction for advanced bonding applications, where bare Cu and PCC wires have known limitations.

Palladium coated Copper (PCC) bonding wire is a popular low cost alternative for Gold wire. The quality of an innovative Pd coating technique on Cu (Direct PCC) wire is evaluated by comparing free air ball (FAB) deformability and stitch bond strength to those of Cu wire. The coating is applied to commercial Cu wire which is either thermally treated or as-delivered. FABs of 40 μm diameter are produced in two different shielding using 20 μm diameter direct PCC (DPCC) and bare Cu wires. The FABs are deformed under a force of 400 mN. The DPCC FAB made in forming gas deforms 0.60 to 1.92% more than the Cu FAB, indicating a slightly softer bonded ball. The stitch bonds are made on Ag of a QFN substrate and on Au of an BGA substrate at 200 °C and 175 °C, respectively. The stitch bond strength of PCC wire is 11–26 % higher than that of Cu wire.

Recently, copper wire bonding has been given increasing attention due to its low cost and excellent material performance. Compared with pure copper wire, Palladium-coated copper (Pd-Cu) wire gets more extensive application partly due to the Pd-coated layer protection for Cu FAB from oxidation in the bonding process. Therefore the research of Pd coverage on free air balls (FAB) is an important aspect for Pd-Cu wires. FAB of Pd-Cu wires may have exposed Cu region on top side of FAB where it is to be bonded, this Pd uncovered part could be possibly extended and then exposed to the outside of bonding area. After mold encapsulation, the exposed Cu region on FAB may have higher risk to be subject to corrosion compared with the Pd covered region and lead to failure of first bonding. This paper presents a study for Pd coverage issues on FAB of 0.8mil Pd-Cu wires. A chemical etching method with FeCl <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> solution for FAB was adopted to analyze the Pd coverage on FAB. The etching condition had been optimized through careful examinations on the surface morphology of etched FAB. With various FAB forming process parameters, e.g., electronic flame-off (EFO) current intensity and heating, Pd coverage has been studied. The Cu exposure, before and after FAB formation, was also compared and analyzed. This analysis results will provide an important reference for the optimizing FAB process parameters in Cu wire bonding process.