Electroless and Immersion Plating Process towards Structures and IMC Formation

The selection of surface finish on printed circuit board is very important in soldering process because it can prevent the copper from oxidation and provide solderable surface. It is also will influence the cost, manufacturing, shelf life, surface quality, final product reliability and environment.Electroless and immersion plating process has now become an important and popular surface finish in electronic industry due to simple and low cost process. The application of electroless plating are widely used in electronic industry, however it also was used in various industries such as aerospace, automotive, oil and gas, chemical processing where the need on the complex shape, homogeneous and uniform layer are required. Besides, every surface finish has their strength and weakness. Immersion silver is one of the surface finishes frequently used on the printed circuit board due to their characteristics such as good solderability and wettability, easy assembly and low cost. Both of these two processes were acts as barrier layer between copper substrate and solder balls. This review was discussed about electroless nickel and immersion silver plated on copper substrate, coating process and parameters involved. The topic also covered include the morphology and thickness of the intermetallics formed during soldering as well as the isothermal aging process. Keyword – Electroless plating, Electroless nickel, Immersion silver, Intermetallic compound, Reflow soldering


I. INTRODUCTION
The selection of surface finish on the printed circuit board could be the most important decision in electronic application. Currently, surface finish is used to protect the exposed copper surface from the oxidation when exposed to the environment which can reduce the strength of the solder joint and the wettability [1]. The electroless and immersion are widely used in electronic industry because both of them can enhanced the bond pad properties for electronic packaging [2]. The different between the electroless and immersion plating process are an auto-catalytic reaction [3][4][5][6][7] and displacement process [7], respectively. Electroless plating is a chemical reduction process [4,5], where the deposition process of the metal was not involved the electrical energy usage [6,7]. Typical surface finishes commonly used for electroless plating technique are electroless nickel/immersion gold (ENIG) and electroless nickel/electroless palladium/immersion gold (ENEPIG). In electronic industry, electroless nickel (EN) is a popular surface finish compared to other techniques such as electroplating. Many researchers have been issued about the advantages of the EN for the engineering and especially for printed circuit board (PCB) application [3,5,[7][8][9]. Several advantages of EN include the quality of deposit such as physical and mechanical properties. There are some of the properties were usable as solderability, high hardness, magnetic properties, resistivity and low coefficient of friction. However, most applications of the auto-catalytic are depend on their wear and corrosion resistance [3,7]. The desired properties can be a range by choosing different temperature, pH value and composition of the bath [3] as shown in Table 1. The immersion plating is a displacement process where the metallic ions in the solution will be reduced on the surface substrates [7]. Immersion silver (ImAg) is one of important surface finish for the electronic assembly and it was Restriction of Hazardous Substances (RoHS) compliant. ImAg also known as an alternative to HASL [10][11][12] and ENIG [13,14] because of their good solderability, wettability, co-planarity and bonding performance [10,15]. According to Wang et al. [10], the selection of immersion silver as a final surface finish is a good decision because it can eliminates the embrittlement of Au-Sn intermetallic layer (IMCs) as well as provides simpler operation and lower cost. Furthermore, immersion plating are not required to use reducing agent because the base material can behaves as reducing agent compared to electroless plating process [16]. In this review, the present work aims to study the electroless and immersion plating process towards structures and intermetallic compound (IMC) formation including the coating process and its parameter, interfacial reaction between solder and substrates.

A. Introduction
Electroless nickel (EN) is produced from chemical reaction which involves nickel salts and reducing agent. The properties of electroless nickel are depending on the quantity of phosphorus in the chemical solution [6,17]. Generally, electroless plating process comprising a source of metal ions, complexing agent, reducing agent, stabilizer, buffering agent, wetting agent, controlled pH and temperature [6,7] as represented in Table 2. EN is a continuous process of metal deposition. Uniform thickness [5] and thick layer is obtained (between 3 ~ 6 µm) as long as the metal substrates still in nickel solution [6]. By controlling the chemical reduction of nickel ion, it is necessary to change the metal ions to the metal in order to form a coating layer on the substrate [7]. Besides that, if the parameter is not properly controlled, the unintended results will be happened because the reduction of nickel ion can take place in all solutions. The electroless plating process usually used to deposit metal or its alloys such as nickel, palladium, silver, gold and etc. to the substrate. Besides, either the nickel containing phosphorous or boron also included in this category [18]. The additional metal elements used in electroless deposition process known as poly alloy [9,19]. Some of binary, ternary and quaternary diagram are referred as poly alloys. For example, binary alloys have Ni-P [19] and Ni-B [20], while ternary alloys have Ni-P-B [20], and Ni-W-P [21]. For quaternary alloys have Ni-W-Cu-P [9,22]. Tables 3 represent the types of electroless  metallic alloy coatings and their features. In additional, electronic micro-component and packaging need a convenient barrier layer between Cu, Ag, Al, Ni, Au, Pd and Co [7], as well as solder to perpetuate a long period of services. However, electroless nickel are found to be a suitable metal which can act as a barrier and protect the substrate due to it slowest pace rate dissolution of the solder and growth rate of the intermetallic compounds formation [24][25][26][27][28] compared to silver, gold or palladium. There are five types of EN coatings such as pure nickel, black nickel, electroless composite electroless nano-coatings and electroless nickel alloy coatings [7]. EN alloy coatings may consist four types, as follow [22]: (1) Acid bath: Ni-P alloy, 3-5% P (low), 6-9% P (medium), 10-14% P (high).

Stabilizer
To stabilize the bath from decomposition by shielding. For examples Pb, Sn, As, Mo, Cd, thiourea, etc. [6,7] Buffering agent Sustain the pH for long time. For example sodium salt and to choice buffering agent depends on pH range used. For pH adjustment such as sulphuric and hydrochloric acids, soda ammonia [7,17,23].
(1) Sodium hypophosphite The most normally used to reduce agent was sodium hypophosphite. It is mainly used for electroless nickel plating because their characteristics can produce a nickel-phosphorus as a coating layer.
One of the powerful reducing agent was sodium borohydride, where avai1able for the electroless nickel plating. Nickel boride will formed because of the presence of nickel ions. Borohydride is catalytically decomposed if the pH value above than 13 for alkaline solution and the reaction product is principally containing nickel. In general, 5 kg of hypophosphite is required while 0.6 kg of sodium borohydride is needed to reduce 1 kg of nickel [7]. However, complexing agents such as ethylenediamine must be used to prevent nickel hydroxide precipitation [3]. (4) Hydrazine Using reducing agent such as hydrazine can produce pure electroless nickel deposition, which is contain up to 90% of nickel [7], with oxygen and 10% of nitrogen [22]. For the example, nickel chloride hexahydrate contain 4.8gL -1 , 32gL -1 of hydrazine and 4.6gL -1 of sodium tartrate dehydrate at 10 pH values and 95 o C of deposition. (also known ration usually up IV element [6].
or joining of s f solution's p he common ac dipic [17].
Process [23] fect on deposi P content, shif tion. e involved y of final also will ng process rd.
ating onto of coating can obtained [29]. The content of phosphorous in EN can influence the behaviour of the physical, mechanical and corrosion resistance properties of the coating. Oduoza and Khan [30] mentioned that the phosphorous content in EN can give effect on the internal stress because the existing internal stress depends on the microstructure Ni-P deposited. The parameter such as pH and temperature also can influence the rate of Ni-P deposition. It is an important to control the deposition rate of Ni-P or bath composition because when the pH decrease, the rate of deposition also decrease [31]. Table 4 shows the effect of pH change on EN plating process. According to Baldwin et al. [32], the temperature of plating bath will increased to the normal operating range when the content of phosphorous is decreased. Due to their properties, the phosphorous content may easy to control. The value of temperature also influences the rate of deposition. When the temperature increase, the plating rate also exponentially increases [23]. The increasing temperature more than 90˚C make the pH value difficult to control [31,5]. The growth structure of electroless nickel by using hypophosphate as reducing agent can be amorphous or liquid-like [18,33]. The amount of Ni-P can determine the microstructure of the coating either crystalline, amorphous, or both combination of microstructure [36]. Currently, electroless nickel process at low and medium phosphorus level has a mixture amorphous and microcrystalline nickel. While, the structure will be fully amorphous when the phosphorus content is high [7,35,36] as showing in Fig. 1. Low Ni-P has created nodular shape ( Fig. 1(c)), where it happened from combination granular type with grain boundaries. The grain boundaries affected the efficiency of low P as a diffusion barrier. These grain boundaries can be eliminated by using high P content nickel [24]. Schlesinger [22] also mentioned that the higher sizes of the nickel crystallites making up when the phosphorus content is lower. Thus, phosphorus can act as an inhibitor of crystal formation. At the temperature above 220-260˚C, the structure of Ni-P begins change and the deposits starts to crystalline and also lose its amorphous structure. Fig. 2 shows the microstructure of electroless Ni-P plating. Besides, the morphology of the structure is cauliflower-like and crystalline structure. Based on low phosphorous content, these coating have better wear resistant and excellent corrosion resistance in concentrated caustic soda. The abrasion resistance and corrosion protection are better enough as well as economical in plating bath work, if medium phosphorous was used. For the high phosphorous, the coatings have good corrosion resistant, very ductile [39], low porosity, low internal intrinsic stress and non-magnetic is as plated state [23].

Solder Joint Microstructure of Ni-P coating
During plating process, intermetallic compound (IMC) will never form because the phosphorus atoms are trapped between nickel atoms in random way. It is because, they can lessen the possibility of contact among nickel atoms [18,38]. After the plating process, the intermetallic layer will be formed as Ni 3 P within the alloy. Ni 3 P is a first intermetallic layer that forms after plating process. This formation will happen when the temperature raises 320˚C as well as the crystallized structures will reach a maximum after heating at 400˚C for 1 hour. Before the formation of Ni 3 P phase in stable condition, NiP 2 and Ni 12 P 5 can be form with medium and high phosphorous contents [7].
From the previous researchers [39,40], the polycrystalline columnar structure of Ni 3 P has a contain defects. Kirkendall voids are defects that form on P-rich layer formation and these voids will be affected the joint reliability that cause brittle fracture. Alam et al. [43] reported that the content of P and thickness Ni-P layer probably effect the performing shear test. Normally, a new layer such as Ni-P rich is formed between interfacial IMC and pure electroless Ni-layer during reflow process [44].
Normally, the type of IMC that formed after reflow soldering between substrates and solder balls was (Cu,Ni) 6 Sn 5 . The formation of intermetallic compound between solder and Ni-P are depending on Cu concentration. If the Cu concentration in the solder was lower than 0.2 wt%, only (Ni,Cu) 3 Sn 4 appeared at the interface. Meanwhile, when the high concentration of Cu is more than 0.6w% the IMC layer was (Cu,Ni) 6 Sn 5.
However was dete between after refl interface interface structure To achie operation influence process a phosphor and reflo layer is a of IMC concentra process.

D.Nickel
The ad finish be resistance significan ability of interconn borane w with Ni-P In genera boron co cauliflow was grow the effec r, when the C ected [24 , 43-4 lead-free sold low soldering will be chan between  Fig. 4. The y. According t sion layer as i 0.2 wt% and investigated a hey found tha eported by [2 n [45]. Fig. 3 and Sn-4.0Ag ype and boom -4.0Cu solder afte and f) low-P [24] w (a) cross sectio Ni-P depositio ameter and c iform layer. T mine the crys ructure and IM ction between cess regardles he type of l e normally IM Ni-B) has gain its ability to erability [

Process Function Cleaner
To clean the copper surface before the next process. Removes surface oils, oxides, any organic material and to ensure the copper surface will be in good condition and to be uniformly micro-etch.
Micro-etching To produce a plating surface that promotes good deposit adhesion. Remove any chemical contaminant and metal oxides by lightly etching the exposed copper surface. The example of etchant types are sodium per-sulphate, peroxide / sulphuric.

Pre-dipping
To prevent any chemical residues from the previous rinsing step and remove any surface oxidation that may happen in the previous rinsing stage.

Immersion silver deposition
To deposits a layer of silver onto all of the exposed copper surface and protect copper from oxidation.

Drying process
To ensure the copper with silver surface finish are completely dry. Remove any leftovers moisture from the board to prevent discolour and to ensure metal quality.
B. Characteristics of the Immersion Silver Surface Finish Immersion silver (ImAg) is one of an alternative finish where it was designed as protective surface finish to ensure the solderability of the underlying substrate [10,15]. The advantages of ImAg finish including good coplanarity, wire bondable [15], and suitable for fine-pitch of electronic applications [67]. According to Barbetta [67], ImAg also has good solderability, and it can maintained through the multiple reflow cycles. Besides that, it is also can maintaining thesolderability or wettability until 12 months before assembly process [12].
During soldering process, ImAg will be dissolved into the molten solder during assembly process because of liquidation of silver into the tin (Sn) based in the solder [10,15] as shown in Fig 7 (a). The thickness of Ag layer is typically less than 1 µm. It is happen because the deposition of Ag process will be stopped when the substrate surface is fully covered with the Ag solution [15]. Furthermore, Cullen [68] stated that the ImAg consist of 0.15-0.55µm thickness layer and it is 100 times thinner than the traditional electroplated silver deposits. In addition, Wang et al. [10] found that in order to get the 0.5 µm Ag layer in range [69], the duration of plating time is around 1 minute to 4 minutes. The Association Connecting Electronics Industrial standard (IPC-4553) [70] stated that the recommended thickness of Ag layer as minimum 0.13µm and typically 0.2µm-0.3µm can be obtained by using 60 x 60 mil pad. The standard also indicated that the possibility to obtain Ag thickness from 0.07µm to 0.12µm is possible, but it may limited for general purpose application only [10,66]. Some researchers reported that the thickness of immersion silver must be not too thick due to brittle solder joint in lead-free soldering, and not too thin to ensure a lifespan of this surface finish during storage [70][71][72].

C. Solder Joint Microstructure of Immersion Silver
During reflow soldering, the interfacial reaction will occur between ImAg and solder. As the result, the layer of intermetallic compound (IMC) will be formed between them and provides a good metallurgical bonding [10]. The type of IMC layer will be formed at the interface is depends on the metal surface and solders were used. Thus, a selection of surface finish and solder type are playing an important role.   [65] w platelet or ne out the micro n. Similar resu erface after ex face substrate om the solder Cu 6 Sn 5 is form terface coppe sed to aging it C layers eithe al. [76] also in epends on the ls. This is bec lder joint; microv 0˚C at 60 s: (a) s substrate [72].
n form at the i [72], the layer sh at 250˚C fo A typical sh Fig. 7 [74]. How e aging and a ened because ate and the Cu or to aging pr form [69]. The e the increasin f the IMC lay lder balls will l saturates qu of solder and IMC  Fig. 9 The formation of chimney at a pathway between caves and silver surface [77].

The Formation of Microvoiding
A small holes (smaller than 50 µm) that form in a solder joint knows as microvoids (champagne voids) [78] as showing in Fig. 8. These voids caused by the unbalanced inter-diffusion ofcopper and tin at the interface between substrates and solder [10]. According to Schueller [79] and Yau et al. [78], the formation of microvoids caused by cavities or empty bubbles in the copper (the source of the microvoids upon reflow) It is happened because of the interaction between silver and copper during silver plating process [11,75,76].
Furthermore, the thicker silver layer can influence microvoid formation [78]. Typically, voids will form at near the surface of the copper pad, where the silver layer was used. The size of microvoids can separated into two groups which are small microvoids for the diameter less than the thickness of IMC layer (about 5 µm), while large microvoids for diameter greater than the thickness of IMC layer (less than 25 µm) [78]. The existence of microvoids can affect the solder joint strength [80], reliability and lead to crack propagation [10]. Besides that, the chemical solution during plating process has to be maintained, followed the specifications and guidelines to prevent the possibility of microvoid formation [80].
Previous researcher has found the "caves" under the immersion silver coating (Fig. 9). The formation of cave happened due to contaminant of inorganic residues material on the copper substrates during plating process [78]. Once the cave will be appeared at the surface, the microvoids formation have high tendency to occur during assembly. Besides that, Cullen et al. [78] discussed that a chimney formation was detected between caves and silver layer. Chimney is a micro pore that connecting the cave to the surface. However, the formation of cave can be reduced by modify the pre-sulphate micro etches solution.
ImAg finish proves to be high performing and low cost and allowed as alternative to both of HASL and ENIG because of a simpler process. Using the immersion process, the layer was uniform and met the thickness criteria and very stable process because the reaction only based on replacement process. It is also a better choice for lead-free electronics application that has good wettability and solderability compared to other lead free surface finishes. During reflow process, the layer of Cu 6 Sn 5 and Ag 3 Sn are growing into the solder. When the solder exposed to the aging process, Cu 3 Sn 4 layer was appeared and Ag 3 Sn platelet disappeared (with the high temperature aging) from the interface of solder. However, the defect known as microvoid and cave formation has been observed at near surface of copper substrate under silver layer coating, however the prevention can be made by maintained and controlled the chemical plating during the immersion silver process.

IV. CONCLUSION
Plating process involves covering outer layer of the copper or PCB board by deposition process. It is act as barriers to prevent corrosion and to improve the surface finish appearance. Every plating process required advantages from process control because it is very important to achieve the desired properties such as physical and mechanical properties as well as to getting quality of deposits. Besides, two crucial parameter such as pH value and temperaturewere affectedthe performance of plating process. This review paper has presented two plating process namely electroless nickel and immersion silver plating including their structures and IMC formation, by various researchers are reviewing the variety of informations.ENhas two types namely Ni-P and Ni-B. Both of Ni-P and Ni-B crystalline structures formation depended on the amount of P and B content. The different both of these content, the different structures will be appeared either the structure are amorphous, crystalline or amorphous and crystalline.While, the ImAg surface finish is an alternative to both HASL and ENIG due to excellent wettability, simple process and economical. There are several types of IMC that formed after reflow soldering for EN and ImAg such as (Cu,Ni) 6 Sn 5 , (Ni,Cu) 3 Sn 4 and Cu 6 Sn 5 , Cu 3 Sn 4 for respectively. The thickness and size of microstructure of IMC layer influenced by reflow condition, solder size and aging time. The increasing of aging duration made IMC layer also increased. Meanwhile, the bigger solder ball sizes produced a thinner IMC layer compared with smallest solder ball sizes where produced thicker IMC layer.