Introduction to adhesives joining technology for electronics
Abstract:
Polymer adhesives are now considered as invaluable materials for modern-day, low-cost miniaturized consumer electronic products because of their low cost and high production throughput. This introductory chapter describes the different types of adhesives that are used in electronic packaging along with a brief introduction to electronic assemblies and the uses of adhesives in electronics.
1.1 Introduction
The assembly of modern electronic components started in the early 1960s, after the emergence of the integrated circuit in the 1950s. However, all the early electronic components were connected by metallic systems such as solder (e.g. tin–lead), and wire bonding (gold and aluminium), and were mounted on ceramic substrates up until the 1980s. In fact, most of the applications for electronic systems at this time were for high-reliability applications such as defence and were not driven by cost. With the introduction of electronics into consumer products, polymeric substrates and adhesives started to attract attention because of their low cost and high production throughput. In fact, polymer adhesives are now considered as invaluable materials for modern- day, low-cost miniaturized consumer electronic products. However, the introduction of polymers and adhesives in the packaging of semiconductors results in a number of manufacturing and reliability problems. Outgassing, ion contamination, void/air entrapment, moisture absorption, CTE (coefficient of thermal expansion) mismatch and related thermal stress are some of the notorious problems that have had to be addressed by polymer experts and electronic packaging engineers over the last few decades. Therefore, polymer adhesive technology (both materials and their processing) has improved significantly during that period. Because of the tremendous growth of the consumer electronics, adhesive technology is now a multi-billion pound business, fuelling continuous R&D efforts from both industry and academia.
Adhesives are used as both functional and structural materials in electronic packaging. They provide electrical connections, thermal paths for extracting heat away from a semiconductor, and can be used to provide structural integrity of the package and system. They are applied as either paste or a solid film. Bare chips, single-chip packages, and multi modules can all be attached on the plastic printed circuit board (PCB) by adhesives without any soldering or wiring. Some of the advanced adhesives are used in flexible PCBs, optoelectronics, and sensors, as well as in smart cards. This book is dedicated to advanced adhesives that have attracted attention from the electronic packaging community. This chapter aims to introduce the various types of adhesives used in electronic packaging, along with a brief introduction to electronic assemblies and the uses of adhesives in electronics.
1.2 Classification of adhesives used in electronic packaging
Adhesives used to package electronic components are classified based on (i) their physical form such as pastes or films, (ii) chemical type such as epoxies, acrylics or polyimides, (iii) molecular structure such as thermoplastic or thermosetting, (iv) curing method such as heat curable, uv-curable, (v) function such as electrical adhesives, thermal adhesives, and (vi) application such as die attach, underfill adhesive.
Interestingly, the same adhesive can be termed differently, based on its various classifications, and sometimes it is desirable to mention all kinds of classification while describing an adhesive. For example, Anisotropic Conductive Adhesive (ACA) is a film-type (physical form), epoxy-based (chemical formulation), thermosetting (molecular structure), heat-curable (curing method) adhesive used in electronics.
While this book is dedicated to such advanced adhesives, a very brief classification of all adhesives used in electronics is given in here.
1.2.1 Classification based on physical form
All adhesives used in electronics can be applied either as a paste or a film. Pastes are semi-solid materials that are usually screen printed or stencil printed. Sometimes, they can be dispensed, even through a needle. Film adhesives are solid tapes that can be cut as required for the bonding area. Isotropic Conductive Adhesives (ICA) are generally used as pastes while Anisotropic Conductive Adhesives (ACA) are applied as films.
1.2.2 Classification based on chemical formulation
Adhesives are much more complicated than metals or ceramics. They are commonly referred to by their polymer type. The most common polymer type used in electronic adhesives is epoxy. The next is probably acrylic, then polyimide and then polyurethane, and so on. There are hundreds of formulations on each generic polymer in the market, with some proprietary additives.
1.2.3 Classification based on molecular structure
Polymer adhesives are broadly classified as thermoplastic, or thermosetting, depending on the molecular structure. Thermoplastics are linear polymers with either straight chains or branched chains (see Fig. 1.1a,b). They melt (liquefy) at a specific temperature and then resolidify on cooling; therefore they are easy to process and rework. Examples of thermoplastics are polyurethanes and polyamides. However, some linear polymer adhesives with aromatic or heterocyclic components in their structures possess high thermal stability and do not melt; instead, they decompose on further heating. Thermosetting adhesives become highly cross-linked polymers on curing (see Fig. 1.1c), and unlike thermoplastics, they will not then melt at any temperature. Therefore, they are not reworkable once they are cross-linked. Examples are epoxies, cyanate esters and phenolics.
1.2.4 Classification based on curing process
There are various curing processes to give workable strength to adhesives. While dispensing the adhesive in its initial state, the molecular weight is low; therefore it is usually in a liquid state. However, during the curing process, molecules join together to form a three-dimensional network, leading to high molecular weight which increases the viscosity and finally forms a solid. Heat-curing is the most widely used and simplest method of polymerization. Curing kinetics are temperature- and time-dependent.1 For any polymer system, the kinetics’ parameters dictate the rate of curing at different temperatures. Final properties of the adhesive are directly related to the curing percentage of the polymer components. The curing profile is measured by monitoring the desired adhesive properties such as electrical resistivity, hardness, bond strength or dielectric constant, as a function of temperature and time. The degree of polymerization, i.e. curing percentage, can be measured by experimental techniques such as DSC (Differential Scanning Calorimetry), Thermomechanical Analysis (TMA) or FTIR (Fourier Transformed Infrared Spectroscopy).
Curing kinetics are mostly controlled by the added hardener (often called the catalyst) used in adhesive formulations. Short curing is desirable to increase the industrial throughput. Therefore, selecting a suitable hardener is important to cope with the mass production required for consumer electronics. Recently, snap-cured adhesives have been developed, which can be cured within a few seconds at relatively low temperatures.
Some adhesives polymerize in the presence of ultraviolet or even visible light, which makes them attractive for wider applications in heat-sensitive parts, especially in photonics packaging. They also show rapid curing. In certain cases, however, slight heat is required to complete the curing process. Chapter 8 describes this photo curing process and their uses in photonics.
Microwave curing is another rapid curing method that is attractive for its advantage in providing a uniform heating capability over a large section. However, unlike conventional microwaves, it uses Variable Frequency Microwaves (VFM) that rapidly vary the operating frequency. Swept frequencies generate a uniform energy distribution as well as reduced arcing of metal fillers and semiconductor dies.2,3
1.2.5 Classification based on functions
Electronic adhesives are used to serve for mechanical connections, electrical connections, thermal dissipation and stress mitigations. As the name implies, the main function of an electronic adhesive is to mechanically bond dissimilar materials with the required strength and maintain that strength over the life-time of the product. Although polymer adhesives are not intrinsically electrically/thermally conductive, their monomers can be wetted by metallic particles and, after curing, those metallic particles are impregnated in their microstructure. The common polymer matrix is epoxy based and silver and nickel particles of 5 to 20 microns in diameter are typically used as the conductive fillers.4 Depending on the metal particle content in the adhesive, they are classified either as anisotropic conductive adhesive/film (ACA/ACF) or isotropic conductive adhesive (ICA). ACA or ACF types contain less filler metal (< 5 % by volume) and therefore can conduct electricity in only one direction. For example, they can be used to provide electrical connections for flip-chip components where the metal particles provide electrical connection between the pads on the die and those on the substrate. The die is assembled onto the substrate by applying the ACA or ACF onto the substrate and then placing the die over the substrate/adhesive assembly and applying pressure and temperature until good electrical contact is made and the adhesive is cured. At elevated temperature, the adhesive is cured or set and the pressure helps to entrap the conductive particles between the bump and the pad; therefore, electrical contact is made only in the z-direction from the substrate to the die. Hence the name anisotropic. Nickel- or gold-coated polymer particles are also used as conductive fillers in ACF.
ICA contains a much higher volume percentage of metallic filler material (such as silver) above the percolation limit, so when they are set by curing, they can conduct electricity in all directions. Hence the name isotropic. Silver-filled epoxies can also be used as thermally conductive adhesives. However, special additives such as ceramics having high thermal conductivity but very low electrical conductivity (e.g. AlN, BN, Al2O3) are used as filler materials where only thermal conductivity is desired and there is a risk of short circuiting. Recently, carbon nano tubes have been used as filler materials because of their high thermal conductivity.5 Thermally conductive adhesives are extensively used in power electronic devices and heat sinks. Chapter 2 illustrates the materials, and the conduction modelling and processing of thermally conductive adhesives.
Flip chip on plastic PCB is receiving significant attention in consumer electronics; however, the low CTE of silicon and the high CTE of a PCB pose a significant threat to the thermal fatigue reliability of solder joints. Underfill materials provide an example of how adhesives are used to reduce thermal expansion related stress (see Chapter 5 for the details) and hence thermal fatigue of solder joints.
Modelling is now playing a very important role in deciding what adhesives to use in electronic packaging. Modelling techniques, both at the micro and macro scale, can be used to identify the electrical, thermal and mechanical performance of adhesives. In addition, modelling can help identify optimal properties for adhesives during the packaging process and subsequently during the life of the package in different environments. Chapter 7 gives some details on modelling techniques that can be used for adhesives in electronic packaging.
1.3 A brief overview of electronic assemblies
It is assumed that readers of this book are aware of the basic construction of electronic assemblies. However, a brief overview of such assemblies might assist in recapping where and how advanced adhesives are used in electronics. Basically, electronic products starts from the semiconductor materials such as silicon. Sequential processing steps (such as doping, oxidation and metallization) on a silicon wafer constitute a device, where devices with a specific design can be interconnected to form a functional circuit. The integration of many such circuits or components on a single chip is called an integrated circuit (IC) or IC chip (or die). By this process, thousands of dies can be produced from a single silicon wafer. Of course, there are many types of ICs whose size can vary from 1 mm to 30 mm. ICs are classified by (i) materials and composition, (ii) degree of integration or number of transistor elements, (iii) principles of operation, (iv) manufacturing method, (v) device type, etc. An IC can be part of a single component, such as a power amplifier or a power transistor, or be a fully integrated microprocessor as used in modern PCs and high performance servers or workstations. Theoretically, there is no limit to integrating multiple functions on a single chip, giving rise to concepts such as System-on-Chip (SOC). However, adding more functionality to a single piece of silicon by decreasing down beyond 22 nm node technology is very costly and time-consuming, due to design and fabrication difficulties.6’7 Thus, in addition to continued developments in SOC, System-in-Package (SiP) is gaining interest to help fulfil the requirements of a number of applications. SiP is characterized by any combination of more than one active electronic components of different functionality plus (optionally) passives and other devices such as MEMS or optical components assembled into a single standard package that provides multiple functions. SiP requires expertise in packaging to ensure that it meets its objectives.
Before the advent of ICs, discrete components such as transistors, diodes, capacitors, resistors, inductors, etc. were mounted on a PCB to form a circuit block, which was then connected to other circuit blocks to build a complete functional unit. ICs have enabled the monolithic integration of most of those blocks onto a single chip, resulting in miniaturized products that have become low in cost and high in reliability. However, every individual IC has to be packaged before it can be used. Packaging starts where the IC chip stops. The typical parameters important for IC packaging include the number and distribution of Input/Outputs (I/Os), materials used, power size of the IC chip, and the number of chips to be assembled into a single package. The constraint for any packaging engineer is that the final package needs to be reliable and cost effective.
Packaging a single IC does not generally lead to a complete system for a piece of electronic equipment, since a typical electronic system requires a number of active and passive components. An IC is known to be an active component in an electronic system. Passive components such as resistors, capacitors and inductors are also integrated with the packaged IC to build a system-level package (e.g. the whole ‘electronic package’). Thus, the term ‘electronic packaging’ is used in a broader spectrum than that of IC packaging. The functions of an electronic package are to protect, power and cool the ICs and/or components, as well as to provide electrical and mechanical connection between the ICs and other components, and to communicate with the outside world.
Figure 1.2 shows schematically the hierarchy of an electronic package.6’8 Typically, there are several levels of packaging, starting from chip level (zero level) to module (first level packaging), card or PCB (second level packaging) and mother board (third level packaging). The number of levels may vary, depending on the degree of integration and the totality of the packaging needs. For example, high performance servers might contain a large number of levels, whereas, consumer electronic products may consist of only one or two levels of packaging. However, among all the levels, the first level packaging, i.e. IC packaging is the most critical step for packaging engineers. Most of the adhesives discussed in different chapters of this book are related to IC packaging.
1.2 Hierarchy of the electronic packaging: optical photographs of the real wafer/package/assembly in different levels.
Since, there are so many types of ICs and their packaging requirements vary over a wide spectrum, it is impractical to have one packaging solution for all ICs. To resolve this problem, many types of IC packaging technologies have been developed that vary in their structures, materials, fabrication methodology, building technologies, size, thickness, number of I/O connections, heat removal capability, electrical performance, reliability and cost. In general, based on the methodology used in assembling the IC packages to the PCB, these are classified into two categories:
If the packages have pins that can be inserted into holes in the PCB, they are called through-hole packages (see Fig. 1.3a). If the packages are not inserted into the PCB, but are mounted on the surface of the PCB, they are called surface mount packages (see Fig. 1.3b).
1.3 Board level integration of the packages for (a) through-hole packages and (b) surface mount packages.6
The advantage of the surface mount package, as compared to through-hole, is that both sides of the PCB can be used, and therefore a higher packing density can be achieved on the board. Dual-in-line packages (DIPs) and pin grid arrays (PGA) are two common through-hole packages (see Fig. 1.4). In DIPs, the i/Os, or the pins, are distributed along the sides of the package. To achieve higher I/O connections, PGAs are used where the pins are distributed in an area-array fashion underneath the package surface. However, most of the advances and varieties are observed in the surface mount packages. The small outline package (SOP, see Fig. 1.5a) is the most widely used package in modern memories for low I/O applications because of its extremely low cost. The quad flat package (QFP) is an extension of the SOP with larger I/O connections (Fig. 1.5c). Both the SOP and QFP have leads that can be attached to the surface of the PCB.
1.4 Optical images of two common through-hole packages. (a) Dual-in-line packages (DIP) and (b) pin grid arrays (PGA). (Courtesy of Motorola.)
1.5 Optical images of three common surface mount packages. (a) Small outline package (SOP); (b) quad flat package (QFP); (c) ball grid array (BGA) package.
In the late 1980s, packages with solder balls were developed as an alternative to packages with leads.6’8 The solder balls can be placed underneath the surface of the package in an area array and significantly increase the I/O count of surface mount packages. A ball grid array (BGA) package is an example of this technology (see Fig. 1.5c). Smaller, thinner and lighter packages are required in this modern age of portable and hand-held products.7 Micro BGA and/or chip scale packages (CSPs) have been developed to address these demands of modern electronics.6 For example, a DIP of the 1960s was roughly 100 times the size of the IC die. Since then, the evolution in packaging technology has reduced the ratio of package area to die area by a factor of 4 to 5. It is important to note that the package sizes are progressively approaching the size of the chip. The CSP, by definition, is a package whose area is less than 1.2 times the area of the IC it packages.
1.4 Typical uses of advanced adhesives in electronics
Figure 1.6 shows a schematic cross-sectional view of a wire bond BGA package with Cu heat spreader. Three adhesive materials have been used in this package, while molding resin has been used to encapsulate the chip and wire bonding. Among these three adhesives, the thermally conductive adhesive that bonds the chip to the heat sink is considered as an advanced adhesive. Details of this adhesive can be found in Chapter 2. In some applications, wire bonding is replaced by ACF or ACA; however, a different design to that shown in Fig. 1.6 is used.
Figure 1.7 shows two examples of direct bonded chip-on-board. The first one is bonded by flip-chip solder joints (Fig. 1.7a) while the second one is bonded by anisotropic conductive adhesives (Fig. 1.7b). To reduce the thermal expansion related mismatch between the chip (3 ppm) and the substrate (18 ppm), the space between the solder joints is filled with an adhesive known as an underfill material. Chapter 5 describes the underfill process for flip-chip applications. On the other hand, it is clear from Fig. 1.7b that ACA itself occupies the room between chip to substrate.
1.5 References
1. Chan, Y.C., Uddin, M.A., Alam, M.O., Chan, H.P. Curing Kinetics of Anisotropic Conductive Adhesive Film. Journal of Electronic Materials. March 2003; 32(3):131–136.
2. Bailey, C., Tilford, T., Lu, H., Desmulliez, M.P.Y., Design for Manufacture and Reliability of Polymer Based Electronics. 2nd International Conference on Polymers in Defence and Aerospace Applications. Hamburg, Germany, 2010.
3. T. Tilford, S. Pavuluri, C. Bailey and M. P. Y. Desmulliez, ‘On Variable Frequency Microwave Processing of Heterogeneous Chip-on-Board Assemblies’, Proc. International Conference on Electronic Packaging Technology & High Density Packaging (ICEPT- HDP 2009), Beijing, China, ISBN: 978-1-4244-4659-9, pp. 927–931.
4. Uddin, M.A., Alam, M.O., Chan, Y.C., Chan, H.P. Adhesion strength and contact resistance of Flip Chip On Flex (FCOF) packages – Effect of curing degree of anisotropic conductive film. Microelectronics Reliability. March 2004; 44(3):505–514.
5. Felba, J., Falat, T., Wymyslowski, A. Influence of thermo-mechanical properties of polymer matrices on the thermal conductivity of adhesives for microelectronic packaging. Materials Science – Poland. 2007; 25(1):45–55.
6. Tummala, R.R. Fundamentals of Microsystems Packaging. McGraw-Hill: New York; 2001.
7. Tummala, R.R., Madisetti, V.K. System on chip or system on package? Design & Test of Computers, IEEE. 1999; 16(2):48–56.
8. J. H. Lau, Ball Grid Array Technology, McGraw-Hill: New York, 1995002E.