“You are what you eat” is an expression that has appeared as the title of a book and a TV series, and is frequently used when promoting “healthy living” from many different perspectives. Similarly, electronic products are complex combinations of the materials from which they are formed, and it is important for us to understand at least some basic aspects of the wide range of materials that are involved. Some of these are incorporated in the final product; others play essential roles during the manufacture of the product itself and of the components from which it is assembled.
As explained in the Module overview, this is the first of two Units containing underpinning information. In consequence, this main Unit text is very much an overview, each main section having links at several levels to supplementary material. It is most important that you make sure that you are familiar with the concepts. As well as links given in supplementary information boxes, there are direct hyperlinks to explanatory texts, and there is always the Glossary to consult.
After a general introduction to material properties, this Unit describes three main groups of materials used, polymers, metals, and ceramics/glasses. In the final three sections, we see how the materials are applied within the context of components, the board, and the materials used to solder the connections.
The choice of materials will depend on the nature of the product, and in particular its application environment and the customer expectations for its reliability. But the designer is rarely able to exercise a free choice – apart from considerations of cost and availability, the decisions made have to take into account the mutual compatibility of materials and structures and the influence that the material choice will have on the processes used, in particular the techniques used for joining and coating the materials.
Elementary chemistry recognises three main “states of matter”, gases, liquids and solids, three groups where the properties within each group are comparable, but between which there are normally substantial differences. All three states of matter play a part in an electronic assembly:
Changes of state are also important in electronic assembly, both the melting seen in solders and during flux activation, and the solidification that takes place due to processes such as cross-linking (in polymers) and crystallisation (from liquids). In many cases, further changes continue at a slow rate even after solidification appears to be substantially complete. Examples of changes that occur in the longer term are creep and solid-state diffusion. And it is important to remember that changes in the form of the material, in its crystal structure, and in its surface, may have a significant impact on the reliability of the assembly.
For those who would like a video lecture introduction to the basic physical principles underlying the gaseous, liquid and solid states, complete with models and demonstrations, we recommend http://www.vega.org.uk/series/vmc/vmc1/index.php. This part of John Murrell's States of Matter Workshop also looks at phase changes and subtle features involving intermediate phases such as liquid crystals, supercritical fluids and pseudosolids.
Of the solid materials that form the bulk of the materials content of an electronic assembly, we can distinguish a number of groupings of materials, based on their structure and characteristics. Of these, the simplest is the group of metals. What do you understand by metal? If you are fixated on the web, you will probably type define:metal into Google, which will yield some surprising results, ranging from popular music to road foundations to heraldry! All scientific definitions fall into two categories:
But few of the metals that we use in electronics are pure elements, with the exception of the base foil used in a printed circuit boards (the layers subsequently plated on the board have many impurities). Sometimes the metals we use are alloys, mixtures containing two or more elements (some of which may be non-metallic) which dissolve into each other in the molten state. Of course, the structure of the material when it solidifies will depend on the materials and the composition. It will also be influenced by the rate of cooling, and the resulting structure will have major consequences for the mechanical properties of the metal. We will be looking later on at the uses of metals, and the key considerations for each application.
All metals conduct electricity as well as heat, so all electronics will depend on having insulators as well as conductors. Key materials used for insulation, additionally providing protection and structural strength, are the very wide range of polymeric materials, commonly referred to as “plastics”. Large-chain molecules based on a backbone of carbon (less frequently silicon), there are only a relatively small number of broad groups of basic polymeric resins. However, these can be combined, and are generally used “compounded” with fillers to reduce cost and modify their characteristics. In consequence, plastics occur in a bewildering variety.
Another type of insulating material found in electronic assemblies is a group of inorganic materials referred to as ceramics and glasses. Created by high temperature processes, these have varying degrees of crystallinity. At the extreme of the spectrum, ceramics are crystalline in form, with a minimum of inter-crystallite material as a binder, whereas glasses have little short-range structure, and are effectively super-cooled liquids.
Whilst the materials considered so far are either good conductors or good insulators, making appropriate modifications to the structure can result in materials with intermediate values of resistivity. Typically these are polymers modified by the addition of conductive fillers, but a few are intrinsically conductive.
Also occupying the intermediate ground as regards conductivity are the important group of semiconducting materials which form the basis of active devices. Although the earliest experiments were carried out using germanium, most of the, diodes, transistors and integrated circuits on which electronics depends are built around a silicon die. However, microwave devices, lasers and LEDs are frequently made from compound semiconductors, such as GaAs or GaN, and there are still some residual uses for photosensitive materials such as CdS.
All these semi-conducting materials have properties that are highly sensitive to extremely small (parts per billion) levels of impurity atoms, vary significantly with temperature, and exhibit a degree of sensitivity to light. In this module, we will not be considering the complex fabrication processes involved in making the semiconductor, although we will be dealing with the mostly mechanical processes involved in making connections between the semiconductor and the outside world.
Almost all the materials we have described so far exist in only a single form, although tin has two allotropes, as anyone who has suffered from tin pest will know. However, carbon is an element that exists in no fewer than four allotropes, with significantly different mechanical and electrical properties, and three of which are used in electronic assembly. Three of the forms are crystalline:
The third crystalline allotrope of carbon was only relatively recently discovered. Named buckminsterfullerene, after the architect who created the geodesic dome, which its closed cage containing 60 carbon atoms resembles, this is the smallest of a set of similar structures. Other more complicated cages exist, and work on the fullerenes continues as part of nanotechnology research. Visit the University of Bristol web site for more information on carbon allotropes.
Amorphous carbon is formed when a material containing carbon is burned without sufficient oxygen for it to convert completely into carbon dioxide. Known among other names as “lamp black” and “carbon black”, finely-divided amorphous carbon is used to make inks, paints and rubber products, and as a conductive filler for plastics; carbon-filled composites using binders can be pressed and used for applications such as commutator brushes and resistors.
The properties of a material depend at least in part on its internal structure, and different types of material occur in different forms, as we have seen most noticeably in the case of carbon. Most metals have crystalline structures, where a relatively simple atomic layout is repeated in a regular pattern.
The distance over which this ordering is effective will depend on the cooling conditions, most metals solidifying in ‘grains’, where the structure is ordered within the grain, but the structure breaks down at the ‘grain boundaries’. However, careful attention given to the level of impurities and the cooling process can produce long-range order, and this is an important feature of the single-crystal silicon used for semiconductors.
In some cases, notably materials such as ceramics, the crystal dimensions are quite small and there is a continuum between these micro-crystallites and the ‘amorphous’ (from the Greek for “without form”) nature of some non-metals.
With polymers, not only is the structure different because the molecules are considerably larger, but there is little order to the molecules. However, there is a long-range structure imposed, at least with thermoset materials, by the linking between molecules that takes place during the curing process. This creates what is effectively a giant molecule, whose boundaries are the outside of the component.
Not only do materials have different internal structures, but they are also used in different formats. For structural components, different solid shapes can be produced by casting, by moulding (for polymers) or by a variety of mechanical means (for metals). Particularly in components, electronic applications also make use of fibres and films (polymers) and wires and foils (metals). These do not necessarily have the same physical characteristics as the bulk material from which they are made, either mechanically or electrically. This is the result of the greater surface area, and of differences between surface and sub-surface structure and properties.
A still greater surface area can be produced when the material is in the form of powder, and we will see a number of applications for powders, either as elements in the processing or as the route to creating materials with a sponge form. We will see the latter both in composites used for packaging and in tantalum capacitors.
Many of the materials used are relatively dense, and attempts have been made to reduce weight whilst retaining desirable characteristics. The resulting forms include foams (open cell or closed cell) and ‘honeycomb’ structures. Whilst the manufacturing methods are different, both incorporate air within the structure in order to reduce the effective overall density.
In many cases the materials used for electronic applications are complex materials with more than a single component. The aim is generally to improve performance, for example by combining strength or lightness, or by controlling the CTE. Using different materials allows them to complement each other’s strengths and eliminate their weaknesses. These complex materials may take the form of alloys, in the case of metals, but many engineering materials used come under the generic category of ‘composite materials’ (‘composites’ for short).
These are materials made from two or more components. One component is often a strong fibre that gives the material its tensile strength, whilst another component (the ‘matrix’) is a material such as a resin that can bind the fibres together and make the material more rigid.
In fact, composites with a polymer matrix, where the polymer is reinforced by fibres of glass, paper or Kevlar, are so common that such materials are frequently all that a designer has in mind when using the term “composite”. However, there is a range of composites with a metal matrix, and reinforcement with fibre or ceramic. A common example is the AlSiC used in packaging for improved thermal management.
Another style of complex material combines multiple materials in a way as to retain the individual elements. One obvious example is the aluminium-coated copper used in making some printed circuit boards, or the copper-sheathed aluminium bus-bars used for power distribution. A more common form of layered material is the multilayer laminate, found both in components and such applications as surface coatings for enclosures, printed circuit boards and SMART cards.
There are significant differences in the way that the layers are joined:
[ back to top ]
Many of its properties are intrinsic to the material, and not related to its form. An example is chemical inertness, although one has to be careful to distinguish between genuine non-reactivity and a lack of reaction due to a coating such as an oxide. For example, the oxide on an aluminium surface is sufficiently resistant to give protection against many common reagents that will react significantly with the underlying metal if it is exposed, for example by accidental scratching.
Whilst surface properties such as insulation resistance and roughness are significant in many applications, for most purposes the bulk properties fall into three categories, mechanical, thermal and electrical. Important mechanical properties are strength, hardness and ductility. Depending on the application, key thermal properties include CTE, conductivity and melting/softening behaviour.
Note that only pure elements or compounds can be relied upon to have a single melting point; alloys frequently have a pasty range, and polymers will either soften over a temperature range (for thermoplastics) or else merely decompose as a result of exposure to high temperature over time (for thermosets). During the melting/softening transitions, the mechanical characteristics of the material change, and so do both thermal expansion and electrical parameters. Many polymer materials undergo a glass transition, which affects all parameters, and has implications both for assembly processing and reliability.
We have already indicated that changes in properties take place at elevated temperature, but the effect is not usually an abrupt transition. Also, real materials exhibit both non-linearity and anisotropy – in other words, their properties are not the same under all conditions and in all directions. Firstly, conductivity and other properties of materials vary with temperature, and that variation may not be linear. For example, the tungsten filament of a lamp may change in resistance only a relatively small amount at around room temperature, but the hot resistance of a lamp can be 10 times that of the cold filament1. And water first contracts when you cool it, and then expands, having its maximum density at around 4°C.
Secondly, there may be differences in all kinds of characteristics depending on whether one is measuring through the thickness of a material or across its surface. For example:
These differences with axis are referred to as anisotropy, as distinct to the materials being ‘isotropic’, that is the same in all directions (from the Greek).
Particularly in the case of complex materials and those made up of a number of different materials, the shape of the material and the way that it has been shaped can both affect its properties. The reason lies in the stresses that have been built into the material during the processing. Some of these will be in the surface layer, as in the case of glasses, whilst others will be internal, as a result of the non-uniformity of the material and the way that stresses have become “frozen” during processing. These stresses can have an effect on the flatness of the material, on its stability during assembly processing, and on its ability to withstand a mechanically severe environment.
Whilst the stresses built into a material affect its properties, for most materials a much greater impact on performance comes from imperfections, such as defects and flaws, whether these take the form of surface scratches or contamination, or boundaries and voids between elements of the substructure. A well-known example is the improvement in glass bottle strength that was produced by coating the inside of the bottle with a thin adherent layer of tin oxide. This reduced the likelihood of crack propagation from defects, and allowed the glass to reach a higher percentage of its intrinsic strength. The result was that the bottle could be made of thinner glass, saving material, cost of manufacture and transport costs.
[ back to top ]
‘Polymers’ is a broad term, covering most materials which are not metals, ceramics or glasses, but which are often referred to in everyday terms as ‘plastics’.
Before you read any further, consider a product made by your company with which you are familiar, or some other electronic product which you either know well or can examine in detail, and try and list as many examples as possible of the ways in which polymers are used. Try to group them into categories according to their physical characteristics: you may expect to find applications for thermoplastics as well as thermosets, and for adhesives and rubbers as well as rigid materials.
Start by producing a draft list of the most obvious applications. Then take a good look at a wider range of products, and see if you can extend your list, before looking at our solution. We don’t offer a prize for the longest list, but would certainly like to receive your comments and ideas!
Polymers are a range of materials that are crucial to electronics. They make possible the printed circuit board itself, which consists mainly of layers of reinforced polymer, and are important elements both in component manufacture and in some assembly processes:
In circuit assembly processes, polymer applications include:
Now it’s time to review what you know about polymers, which we invite you to do by first reading the ‘Note on nomenclature’ below:
In the note on nomenclature above, we can identify four groups of key concepts:
Are you clear about all these? If you aren’t, or aren’t sure, then we strongly recommend you to read the two documents at these links:
Polymer basics (Background; Polymer structure; Initiating cure)
Polymer types (Elastomers; Plastics; Adhesives; ‘Tailoring’ polymers)
Polymeric materials vary widely in their mechanical properties, depending on the mix of base resins used and the additives; they can be compounded to be tough, or stiff, and with varying degrees of rigidity. For structural use, they offer comparable performance to many metals, with much lower weight, and usually lower cost. Also a key advantage of a polymer solution is that little finishing is required, since the colouring can be compounded into the polymer, and the surface finish is determined by the moulding.
However, most polymeric materials have limitations in regard to operating temperature which are set by the glass transition temperature of the material or its softening point (depending on whether we are considering a thermoset or thermoplastic resin). Where the elastomeric properties of the material are important, then the user must be aware of the changes in performance produced by the environment and conditions of use.
Another key difference between polymers and metals is that polymeric resins are intrinsically flammable, so flame-retardant additives must be added.
Supporting information on these issues is contained in:
There will be more information about flammability in the references that accompany the discussion on laminates later in this Unit.
Board fabrication is a major user of polymers, as we will see later in the Unit, and of course many of the components we use are polymer-based, as we shall be describing in Unit 3. However, many assemblies make use of elastomeric materials and adhesives. The requirements for those adhesives that are used to attach SM devices to the underside of a mixed technology board are stringent, and there is significant information on this topic in our paper on Gluing components.
When glues are used for component attachment, the only aspects of their thermal performance that matters is an ability to withstand temperature shock. However, for other applications within board assembly we deliberately tailor the resin to be thermally conductive, although most polymers are relatively poor conductors of heat, however heavily-loaded with alumina. There is more about this in AMI4817, Design for Thermal Issues. Polymeric materials selected for thermal performance are not necessarily adhesives, some constructions preferring to use elastomers, as these have the ability to accommodate surfaces that are less than flat.
As with all our discussion on materials, we would encourage you to look at real applications and become more aware of the range of materials used.
Polymers used for insulation need to have adequate dielectric strength, and be resistant to breakdown, both through the bulk of the material and across the surface. These aspects become important when we are designing a high voltage circuit. But such considerations are not just for specialists – many items of instrumentation need optoisolators to protect them against significant voltage events in the uncontrollable and hostile external environment.
Used as much for their dielectric properties as their voltage withstand, polymers are used in many capacitors, particularly for mains voltage applications, such as filter capacitors. Here you need to be aware of the concepts of dielectric constant and permativity, and the possibility of dielectric loss. And of course the dielectric constant of the laminate used in a printed circuit board becomes an important factor to be considered when designing high speed circuitry. More information on these aspects in our paper on Dielectric properties, and much more about the consequences in AMI4814 and AMI4822.
Most polymeric materials are inherently insulators, requiring the addition of conductive fillers, or of a surface-active agent that will attract moisture, in order to be even a poor conductor. This is one of the factors that makes it difficult to specify materials for ESD-protective bags. However, the insulation properties of polymers are widely seen at all levels of the electronic assembly. For example, some capacitors have a polymer dielectric, and many components possess a polymeric outer package, providing insulation as well as protection from humidity and handling.
Polymers for insulation are frequently solids, but significant use is made of sheets and films: at the enclosure level, a key component using polymeric insulation is the wire cable used for interconnection. Although electrical wires and cables are extremely diverse, with specifications to suit particular applications, simplistically they are arrays of polymer coated wires or co-axial cables within a sheath that gives mechanical and environmental protection:
Polyethylene and PVC are the main materials used for insulation and sheathing, often compounded with a small amount of carbon black to give UV stability. However, PVC has a higher relative permittivity due to its polar structure, and for many applications polyethylene is the ideal polymer. It has excellent electrical properties, is relatively cheap and easy to process, and has a wide range of physical properties which can be optimised by selecting the correct molecular weight and degree of branching. Linear low density polyethylenes, and their copolymers with vinyl acetate are also used to enhance resistance to stress cracking.
The flammability of cable is important, especially in critical applications or hazardous environments. PVC has been widely used, because its halogen content gives it relatively low flammability, and halogenated flame-retardant additives are incorporated in other polymers to reduce flammability. However, once halogenated materials ignite, the smoke given off is frequently dense and toxic, and the gases emitted, such as hydrogen chloride from burning PVC, are corrosive and can seriously damage electrical and electronic equipment near the site of the fire. These problems have led to a much greater emphasis on the use of non-halogenated materials for low flammability constructions.
For cables, one approach is to incorporate a high proportion of fire retardant filler such as alumina trihydrate. Between 160–260°C, this decomposes, absorbing heat and releasing up to 30% of its weight as water. The filler suppresses ignition and flame spread by absorbing heat and excluding oxygen.
A parallel development has been in fire-retardant and high-temperature resistant ‘high performance’ polymers, such as polyetherether ketone (PEEK). These combine thermal stability with excellent all-round physical properties, but need a high processing temperature and are rather expensive, so tend to be reserved for more specialised applications.
This development in materials for cable insulation is an interesting example of the way that many different materials may be used to achieve a similar result, but there are always trade-offs, and some of these are unexpected. PVC is easy to work with, but can be dangerous in a fire hazard situation; building a high proportion of any filler into a resin will reduce its workability, and may affect its electrical characteristics; there is always a “top of the market” solution, but this tends to be too expensive for general use, and may have some application problems.
While most polymers are intrinsically non-conducting, resins heavily loaded with conductive particles (predominantly silver) have for nearly 40 years played an important part in semiconductor assembly as a chip bond adhesive, and have received more recent attention as a potential replacement for solder.
The first of these is crucial – almost every semiconductor has a thin bond line of silver-loaded epoxy. However, the jury is still out on the wider use of loaded polymers, with problems associated with rework, the incompatibility of loaded resins and components with fusible coatings, and the difference in flow characteristics. Here solder wetting is a more effective way of spreading across the surface to be coated than wetting by a polymer, unless the latter is very fluid.
The brief on Conductive polymer systems discusses these issues in more detail, and gives an insight into the comparative early success of the anisotropic conductive adhesive, or “Z axis” adhesive, which is often of use for connecting to components such as flexible circuits and liquid crystal displays, which are not suited to solder attach.
What options are available for replacing solder in making connections between a rigid board and a fine-pitch QFP integrated circuit?
Compare your answer with this one.
Unless designed as an open-cell foam, most polymers are hydrophobic, do not wet well, and are resistant to moisture penetration. In consequence, resin coatings are frequently used for humidity protection. That is not to say that polymers will not absorb some moisture within their structure (typically the weight absorbed will be 0.3–1% at steady state under humid conditions), but rather that intimate contact with the polymer prevents any build-up of water next to the sensitive areas of the component.
Now read our short paper on Humidity protection – it’s not about raincoats! – and think about the typical problem posed in our next exercise.
Your latest product includes a number of identical high-voltage circuit elements, mounted on a printed circuit board, and is intended for use in a country with consistent high temperature and high humidity. How might you use polymers to help the product survive its environment? And what would be the implications for its manufacture?
Hint! Don’t restrict yourself to thinking about just a single way of helping the product survive.
So far we still haven’t listed the largest user of polymers, measured by the weight of compound used, which is the enclosure. Typically this will be an assembly of different parts, often made in different materials, even though doing this makes recycling more difficult.
A typical large enclosure separates out the function of structural members from the cladding, and the contribution of polymers at the rack level is restricted to coatings for metals, and “housekeeping” functions such as cable trays, cable ties, connector shrouds, and identification labels.
But a plastic moulding can form a complete enclosure, with integral structural members, and which is self-finished, needing no surface coating to be proof against the average environment. Combine that with the significantly lower cost of moulding against machining, given sufficient production volume, and it is not surprising that plastic enclosures are common for small items of equipment.
[ back to top ]
We are moving now to our second group of materials, the metals. Compared to polymers, there is more evident variety here, and even the casual observer can see differences between the copper used in copper wire and the steel cable used for a bridge, even allowing for the difference in scale. Typically the “man in the street” is used to ferrous metals, such as cast iron, mild steel and stainless steel, a small selection of non-ferrous metals (typically copper and aluminium) and the noble metals, gold, silver and platinum. But that isn’t the end of the wide range of metallic materials used in electronics. If we look at Figure 1, which shows a selection from the Periodic Table, we can see a much wider range of elements that are inherently metallic in nature shaded in grey. Materials that we haven’t mentioned so far, but which are important to the electronics industry are tantalum and niobium (capacitors), tungsten (lamps), palladium, ruthenium (components and surface finishes), and of course tin, bismuth and lead, which are found in solders.
Apart from having a generally “metallic” lustre when clean, all metals have two features in common:
The first of these features has the effect that the “electron cloud” in the metal will drift when an electric field is applied, so that metals are conductors of electricity. Some are better than others, silver being on the top of the pile, and the much cheaper copper quite close. Aluminium is also occasionally used for power cables: though of high resistivity, it is much less dense than copper, so equivalent conductivity can be provided at lower weight.
Metals are also used because they are generally good conductors of heat, though the transfer of heat owes more to the crystal structure than the form of bonding. But of course, even in an electronics product context, most metals are used because of their strength, ductility and general ease of fabrication. They may also be selected for their aesthetic appeal, both architecturally, with the use of copper cladding, and at a more domestic level with stainless steel items.
Metals are such a common part of everyday experience that we take them for granted, without necessarily being aware of their structure and the fact that, like polymers, there are very few pure metals used, but most are alloyed. Even your gold ring will have been alloyed with copper to make it hard enough to withstand normal handling. Not only are the properties determined by the alloy, but also by the history of the material, how it was cooled, how it was subsequently heat-treated, and so on.
All these processes affect the structure at a microscopic level, and the properties of any metal are greatly influenced by the size and shape of the grains, by any flaws within the material, and by any inclusions within the bulk. Some of these inclusions may be materials deliberately added, but others are created by reactions between the alloy constituents, producing local areas of intermetallic compounds (IMCs). Such reactions are not finalised when the alloy solidifies, but continue to change over life as a result of continued solid-state diffusion. Such processes accelerate as the temperature of the metal nears its melting point, so are particularly important for solders, which generally operate at a high homologous temperature.
Fortunately you do not need to be a metallurgist to have a good understanding of product build, but you do need at least some understanding of metal basics and to know what happens when two or more metals are combined. You also need to understand what happens to metals when strain is applied, as the yield and deformation that takes place is the key to manufacturing processes as well as a frequent cause of failure. Information on these topics is contained in these two resource papers:
Metals basics (Metallic properties; The structure of a metal; Alloys)
Mechanical properties of metals (Stress; Strain; Shear strength; Yield and deformation)
These papers remind us that metal structures have grains and grain boundaries, and that the structure has an impact on many aspects of performance, as well as affecting the surface texture. For example, the boundaries affect bonding to the surface, the way the material wets, and the wear that takes place when surfaces come into contact. Make a list of the applications of these concepts to electronics assembly as you read our paper Microstructure of surfaces.
The structure of a material will determine its stress-strain characteristic, and the point at which a plastic strain and deformation take place. It also determines the fatigue strength of the material and the form of failure . Read our paper Stress and its effect on materials for insights into these topics, and two other issues that affect metals, the creep that happens when materials are subjected to continuing stress, and the way in which microscopic flaws can amplify stress at the crack tip, so that cracks propagate and may cause failure. Of course, these aspects are common to all materials, and similar effects will be seen with polymers.
[ back to top ]
Our final generic group of materials contains a wide range of inorganic materials that are generally non-metallic, and in most cases have been treated at high temperature at some stage during manufacture. We have grouped together some crystalline materials, normally referred to as ceramics, with materials that are similar in composition but have been cooled quickly enough to prevent crystallisation taking place, so that they retain their glassy state. As you might expect, there are also materials referred to as “glass-ceramics” which have a high crystalline content, but some glass remaining, and properties intermediate between ceramics and glasses.
We have written a short briefing paper on Ceramics and glasses, which will give you an insight into the value of these materials, which are found more widely than one might perhaps expect.
Based on the descriptions of ceramics, glasses and glass-ceramics given in Ceramics and glasses, and your knowledge of electrical and electronic products, what applications can you identify?
[ back to top ]
If you have worked your way through the last activity, you will already have identified a number of applications for ceramics, glasses and glass-ceramics in component manufacture, and this is just the beginning of a very long list of materials used in making components. Many of these will become clearer as you read Unit 3.
Not only are metals themselves important, but metal oxides play a large role in the manufacture of electrolytic capacitors, where the oxides of tantalum, niobium and aluminium are all important. For this application it is not just the permittivity of the oxide that is important, but also its structure relative to the underlying layer of unoxidised material, the relative CTEs of oxide and underlying metal, and the integrity of the bond between them. [You will recall that iron oxide tends to hydrate, expand and flake off, whereas the oxide layer in anodised aluminium is sufficiently robust to act as a permanent surface finish.]
Another oxide of critical important is silicon dioxide, not as a component of the glasses, but rather as a passivation material used within semiconductors. And of course our list of materials has not so far included any mention of semiconducting materials, of which silicon is the most widely used. There is not space in this module for detailed study of silicon, and even less for the wide range of compound semiconductors, but packaging professionals need to be aware of the general nature of the material. Silicon is hard, brittle and crystalline, and tends to cleave along the crystal plane, rather than at right angles to the surface. Being brittle, the corners of die are prone to cracks and chip-outs, which may have an impact on reliability. The basic material has a low TCE, which affect the design of packages, especially those where the silicon itself forms a major part of the structure, such as the flip-chip.
Because the active parts of semiconductors are created in the surface, that surface itself needs protection. The first layer is of a glass-like material combining silicon dioxide and silicon nitride, but this provides insufficient environmental protection. Alternatives for doing this include hermetically-sealed (“cavity”) devices made using metals, glasses and ceramics, but this approach is expensive. Most semiconductors are now polymer-encapsulated, with a thermoset (usually a modified epoxy) in direct contact with die and bonds. However, in many memory chips, a polyurethane intermediate layer is used, more as a way of soaking up alpha-particles and preventing memory error than for its water-repelling properties.
The bonds of which we spoke are typically made of very fine gold wire, although aluminium is used in some packages and for power devices. The metallisation on the die is typically aluminium, although there is a trend towards copper for this purpose. The sub-surface metallization includes tungsten or chromium, and some device types use titanium and vanadium. Altogether, a most interesting assembly of materials for the materials scientist!
Die bond materials is a paper which examines the wide range of materials used to mount a semiconductor die onto its carrier. Whilst the most commonly used material is a silver-filled epoxy adhesive, there are many other options:
Mind-boggling! You may not need this specific information, but this is a good example of how a problem can be approached from different angles with widely different results.
[ back to top ]
Apart from the enclosure, the printed circuit board is physically the largest component in most electronic products. PCBs are composites, metal-polymer laminated materials of considerable complexity. The majority of those used in professional circles are made with epoxide resin, reinforced with glass fibre, although phenolic resins reinforced with paper are also used, and there is a wide variety of more specialised materials. But the laminate forming the heart of the board is not the only use of polymers in PCB fabrication: the photoresist used for patterning, and applied as a thin film by either wet or dry film techniques, is one polymeric material; another is the ‘solder mask’.
Whilst multilayer boards are assembled by the fabricator from less complex precursors, the starting point will be single-sided or double-sided laminates produced by specialist laminate suppliers. The general principles of manufacture are the same, though there are many differences in the materials used and the final specification of the product. Start your study of laminates by reading our paper on Basic board materials, which outlines the basic process by which laminating resin, reinforcement and foil are brought together to produce a laminated construction for use by the fabricator.
This document confines its attention to the two main laminate types in common use, a phenolic-impregnated paper board aimed at the bottom end of the consumer market, and the main workhorse for profession and multilayer applications, which is an epoxy resin reinforced with glass fibre. Note that the materials choice determines the process choice for both laminate manufacturing the fabrication, because the much cheaper punching process used to create holes being is only compatible non-woven materials. Which is why we find increasing use of composite materials that combine this advantage with the improved characteristics of the epoxy material.
It isn’t enough just to have a sketchy outline of how laminates are made, but the user needs much more information about the characteristics of materials in order to make an informed choice, and specify the right material for an application. The second paper, Properties of laminates, gives an extended account of these issues, and will help you to understand a specification sheet. One key factor to look out for is glass transition temperature – this is a concept briefly explored in the material on polymers, but dealt with in more detail here, because this inherent characteristic of the resin material has an impact on the reliability of through-hole plating and sets a limit on the operating temperature of the board.
The TCE is not the only characteristic of interest, and subsequent sections deal with mechanical and electrical characteristics – these are aspects you should be aware of, although most applications do not stress baords to their limits. Thermal characteristics are perhaps more important, particularly now that designers are cramming more and more power into smaller and smaller space. Whilst the board is a significant heat sink, the thermal resistance of the resin component is high, and strategies such as the use of thermal vias are increasingly necessary.
You also need to understand what happens when a board becomes heated to high temperature, when delamination can occur, with the resulting loss of integrity of the laminate. This is a potential problem that must be circumvented by having appropriate controls on reflow and rework processes, offering another example of the materials and process interaction.
Of course, exposed to still higher temperatures, the board may actually burn, and almost all boards are made to standards that dictate at least a degree of fire retardancy. This is a complex area, with a number of different standards and tests, and particular concern in the light of threats to the continued legality of some of the best materials available for this purpose. Fortunately, in a decision published on 15 October 2005. the EU has exempted deca-BDE, a brominated flame retardant used widely in plastic components, from the RoHS Directive. It would seem that real concerns about fire safety have won at least for the moment, against a concern about long-term health issues.
More on this topic in our paper on Flame retardancy additives and at http://www.bsef.com/bromine/what_the_experts_say/index.php.
The extended trip through laminate properties concludes with a section on quality characteristics, to enable you to understand the work done to control the product, and with information on specifications.
So far we have confined our attention to phenolic-paper and epoxy-glass laminates which come under the generic NEMA descriptions of FR-2 and FR-4, and looked only briefly at the range of desirable properties of a laminate.
Before you read further, we would like you to review and extend your knowledge in this area. Restrict the scope of your survey to the prepregs and laminate (base laminate plus copper) that are supplied to the fabricator – we shall be looking at board finishes later.
Be aware that, in many cases, it may not be possible to get information on FR-2, whereas equivalent information on FR-4 can be found. Don’t worry about this: even our extended searches didn’t provide answers to everything!
Then read our comments.
In the course of the activity, you will have seen that each of these two main runners is actually a grouping of similar materials, rather than a unique specification – materials exist in many different grades, with a variety of characteristics, and a corresponding range of prices. But, while FR2 and FR4 meet many application requirements, they have limitations, and designers need to be aware of a wide range of alternatives. Our paper Alternative board materials gives an account of the “mix and match” possibilities available, using a range of different resins and reinforcement materials. In later sections the document indicates some of the variety of materials available, and especially the issues involved in selecting a high-frequency laminate.
At the very end, there is a section that deals with alternative foils – you may have noticed that we have assumed that all boards are made of copper! This isn’t actually quite true, because increasing use is being made of foils that are themselves laminated structures. However, whilst there are process and performance benefits to these materials, almost all foils present a copper surface to the end user, though this is finished in some way to retain its solderability.
[ back to top ]
Having looked at a range of laminate materials. Now, as we consider a key polymeric material which contributes to the ease of assembly and reliability of the board, we are more fortunate, as there are relatively few materials and processes available.
Rather than just tell you about solder mask, we want you to do some of the thinking!
Base your answers to the questions below on what you know about the assembly process, about polymers, and about adhesion and flow.
Compare your answer with the discussion in this paper.
We hope that in the course of this activity, in preparation for later study, you may also have thought about how one would set about applying such a solder resist, and what might be the design and quality issues about which a designer should be aware. Note that, while we made a comment about solder mask being a “green coating”, this is not necessary the case, and masks of different colours have been used for many years. It is interesting that a recent article3 has highlighted this wider range, so perhaps we will see more “designer boards”!
[ back to top ]
Finishes (or coatings) on the copper pattern of boards are applied to guarantee good solderability, even after prolonged storage, and to provide the required reliability during product lifetime. Ideally, a board finish should give extremely flat pads, a long shelf life without solderability deteriorating, and high solderability even after multiple soldering cycles, whilst being environmentally friendly and cost-effective to implement.
There are two main categories of finishes.
Metallic coatings, which form a wettable surface layer on the exposed copper of the PCB, and are chosen to be much less susceptible to oxidation than the copper on which they are deposited. This group is further divided into fusible coatings such as tin-lead which melt at soldering temperatures, in contrast to non-fusible coatings such as gold.
Organic coatings, which are designed to preserve the wettability of the existing surface by preventing oxidation and contamination.
For many years, the two front runners for the solderable finish on a board have been Hot Air Solder Levelling (HASL) and Electroless Nickel Immersion Gold (ENIG). As we describe in later units, HASL is a process where the board is dipped in molten solder, wets the copper pads, and as much surface material as possible is removed by hot air blasts. When the process is carried out correctly, HASL is an easily-soldered finish with extended life, and has the merit of cheapness. However, the profile of the coating is slightly domed, which has consequences for later assembly of fine pitch components.
ENIG, as the name implies is a plating process, so inherently flat. But it is more expensive than HASL, and requires good process control. Typically HASL finishes were preferred by commercial users, and ENIG finishes found favour for high specification boards. However, there are quality and environmental issues that affect both finishes, and a great deal of attention is still being paid to devising effective alternatives. Drivers for change, especially away from HASL, have been:
Jim Reed of Texas Instruments put it this way as early as 1996: “Faced with a variety of issues, such as environmental (CFC and lead elimination), new packaging technologies (finer pitch devices, Chip-On-Board, flip chip, etc), design (co-planarity), cost (global competition), and metallurgical (solder wetting), future PCB surface finishes will most likely need to be different from the current tin-lead finishes”.
The industry has tackled these issues from a wide range of starting points, and devised a remarkable number of alternative finishes – not only are these different in kind, but there are significant variations in the implementations by different process vendors. However, we are starting to reach some consensus, with considerable interest in immersion tin and immersion silver as alternative plated finishes.
Before you read further, you will find it helpful to review the knowledge of board finishes that you have acquired during your career, and draw up a list of the requirements that one might have when selecting a board finish.
When creating your list, divide it into three groups, considering respectively the issues that are important from the different perspectives of a board fabricator, a board assembler, and a board designer.
Compare your list with our comments.
Before looking at alternative materials, we should examine in more detail some of the elements within this list of issues. For example, the statement has been made that the ideal surface finish should be flat. Why is this important? Draw your conclusions, and then review our comments.
Then it is important that the surface should be solderable, and remain so for an extended period. Solderability is a topic that will recur throughout this module, because joint problems account for around 40% of test failures, and poor solderability is a contributor in many cases.
When we are assessing a board finish, we need to think about how this might be tested for solderability, and how its solderability might be expected to change with time. These topics are discussed in our paper on Solderability.
The search for a flat and solderable finish that can be produced at low cost has been going on for many years, using a range of organic coatings to preserve the solderability of the underlying copper surface, rather than adding extra metal. This has had some success, although Organic Solderability Preservatives (OSPs) are still the choice of only a minority of users. Learn more in our next activity.
Read our paper on Organic board coatings, and consider how current Organic Solderability Preservatives meet the requirements for a board finish?
Review your answer in the light of our comments.
You will have seen that organic coatings, though superficially attractive, have limitations, so plated finishes are generally preferred, though inevitably more expensive. Much work has been done on two immersion plating alternatives to ENIG, using either tin or silver. The processes are conceptually quite similar and the difference in cost marginal, since the coatings are very thin and the bulk of the process cost comes from operation and control, rather than raw materials.
They are also similar in that there are user concerns about reliability. In the case of immersion tin, the concern relates to tin whiskers; in relation to immersion silver, the concern is that silver is a material prone to electromigration. It can be argued from an engineering perspective that these concerns have no consequence, but customers who are concerned about highly visible failures due to electromigration and tin whiskers, albeit in substantially different circumstances, are hard to convince! Read our short briefs on Immersion tin and Immersion silver before tackling the next SAQ.
Which are the main alternative finishes to HASL, ENIG and OSP? What processes are involved in their use? Are there any potential reliability concerns for any of these coatings?
A final comment
HASL and ENIG are not the only finishes on the street! When reviewing the options, you should realise the critical importance of working with both the end-user and the board fabricator to choose the most appropriate surface finish for the application:
The final finish needs to be selected based on design, fabrication issues, solder joint integrity/reliability, availability, and cost.
George Wenger (Celiant Corporation) posting to IPC TechNet on 19 April 2002
What we would like you to do is to draw up a table comparing the materials: this matrix should have the available materials as the heading for each column and the parameters which are important labelling the rows. For many of the entries you will have to make judgements of the sort ‘high’, ‘medium’, and ‘low’ – in other cases you may be able to put figures.
In order to populate your table, you should start by reading what we have to say about each coating and then look for other comments. For example, you could try searching Google for "white tin" +PCB +process, which will give you results which mostly relate fairly accurately to our requirement.
Compare the table that you populate against these.
You should not expect to find that all board suppliers are in total agreement. Part of the reason for this is a preference for certain finishes, based on the company’s experience, the level of local technical support and commercial factors. When you are selecting a board fabricator, then you need to be sure that the finish you ask for is within their repertoire, and is one that they can control – they should be aware of the quality issues and compromises involved and be able to advise you.
[ back to top ]
Not all surface coatings are added to improve solderability, and we have identified two specific cases for consideration, edge connectors and key pads. Ironically, in both cases care has to be taken to ensure that solder does not come into contact with these areas of the board. In particular with edge connections, because of the high wettability of the gold plating and the difference in colour when it is soldered, extreme care has to be taken to avoid even the smallest droplet of solder from coming in contact with the plating.
As a plated ‘gold finger’ on a circuit board, a flash plating of soft gold or immersion gold would have a life only in the range of 1–5 insertions/extractions. However, adding small quantities of nickel or cobalt to the electrolytic plating process produces so called ‘hard gold’, which exhibits much lower wear in sliding operations. Using an under-plate of nickel also improves wear resistance
The number of cycles which fingers will survive depends on the mechanical design of the female connector contacts (geometry and mating pressure), the surface roughness and hardness of the mating surfaces, and the plating thickness. These variables are hard to characterize, so it is difficult to derive a definite relationship4 between their values and contact life. However, it has been found that increasing the gold and nickel thicknesses provides significant increases in connector life: typical values used are 1.5–2.5µm gold over 3–5µm nickel.
Contacts with finishes that have the potential for surface oxidation, such as tin, require a different connector design because of . The most common solution is to have a hard wiping action, but this quickly cuts through the plating, limiting the number of insertions before damage is done to the underlying metal.
Some card edge connectors used to have ‘knife-edge’ connectors which were tin-lead plated and designed to be used with tin-lead plated board edge connections. This provided a ‘gas/air tight’ seal and gave a sufficient number of insertion/extraction cycles over flash coatings.
Note that the types of plating used for edge connection areas are likely to produce unreliable reflow-soldered joints. The combination of maximum gold thickness and minimum solder volume should result in a joint containing not more than 1.4–1.8% of gold by volume (3–4% by weight).
Whilst separate switches give higher reliability, many products use keypad areas on the board which are shorted by pads of conductive elastomer. The on-resistance of each of these switches is relatively high, of the order of tens of ohms but this is generally not an issue, because the switches lead to high impedance inputs.
The key-pad areas on the board are most frequently ENIG, but immersion tin and silver are also used. Although their performance and image definition are not as good, screen-printed pads of carbon and silver-loaded epoxy are to be found on some product.
Apart from areas intended for soldering or wire bonding, what other coatings might you be expected to find on some boards?
[ back to top ]
Most assemblies currently use solder in order to create connections between components and the printed circuit board. Although conductive adhesives can be pressed into service for this application, solder remains far and away the most widespread joining medium. Familiarity, low cost, high reliability and ease of use mean that solder will continue to be a main player.
In this section we are looking first at solder, both the traditional tin-lead alloys and the more recent lead-free types, then at flux, as an essential part of the soldering process, but one where only residues are left at the end of the process. There are of course other ancillary materials, particularly those used in cleaning, but consideration of that topic is in Unit 2. Finally, we will be looking at how bulk solder and flux are turned into solder paste for use in reflow soldering.
Solders have been used since antiquity for joining metals using low-melting alloys that would wet the surfaces to be joined and cool to form a solid joint, without exposing the materials to be joined to a high temperature that would distort or melt them. The original plumbers used a 50% mix of tin and lead to join lead piping and sheet – the Latin word for lead, ‘plumbum’, gives the trade its name; much earlier, alloys of tin with precious metals were used in the fabrication of jewellery. The key to both uses lies in the fact that alloys of two or more metals may have an anomalously low melting point at certain compositions.
We cannot ignore the fact that, for most of the last century, electronics have been assembled using tin-lead alloys, and indeed we can learn much from the study of these materials, even though the latest lead-free alloys are significantly different. To understand the implications of changing materials we can build on many years of experience and much supporting evidence of reliability. Besides, tin-lead has not totally disappeared from the scene, because it will continue to be used for products in exempt industries, and there are applications, such as in high-melting solders, where there is currently an exemption from EC RoHS legislation, granted on the ground that no equivalent material has been identified.
We invite you first to study our paper on Traditional solder materials. This will remind you of what a phase diagram looks like, and the information that can be gained from it in terms of melting temperature, pasty range, and what happens during cooling.
Many of the aspects of this paper are unchanged, such as the potential for under-cooling, and the effect of cooling rate on microstructure. Although tin-rich alloys have significantly different structures, these too will be affected by the cooling rate, and will have a tendency to change and coarsen with time.
Another aspect that is unchanged is the fact that solders operate at high homologous temperatures, so that the bulk material is not very strong, and this influences our decisions on the types of joints to be made. This is particularly important where the volume of solder is comparatively low, as in the joints to many newer, smaller packages.
And the paper indicates correctly that the strength of the bulk material will depend on composition, as does the melting point, and that changing the composition significantly gives us the ability to create solder assemblies that operate at higher temperatures.
Unfortunately, Colin Lea has never revised his excellent book A Scientific Guide to Surface Mount Technology, which is still in print nearly 20 years since its publication. But his information on the rate at which the dissolution of silver into tin-containing solder is affected both by temperature and the presence of silver in the solder alloy, is still valid within the lead-free context, since it has always been the tin in solder that promotes dissolution of metal from the surface being soldered. As we will see later, with lead-free alloys, this dissolution is now more of an issue with copper than silver.
The obvious question, in the light of pressures to remove lead from electronics, is which solder has replaced the tin-lead alloys that have been the mainstay of electronics for the last century?
Detailed consideration of this question is beyond the scope of this module, though you can research this if you want to in Unit 4 of AMI4982 Lead-free Implementation. The short answer is that tin-lead solders have been replaced by a range of high-tin materials, with higher melting temperatures than eutectic tin-lead. Fortunately, it is proven possible to confine the increase in temperature in the soldering processes to a level that components will tolerate. This has been done by improved processed control, although significant work has had to go into this aspect of the change to lead-free.
One unintended consequence of using high-tin materials has been the emergence of concern about “tin pest”, a spontaneous conversion of metallic tin to a powdery allotrope. More information at this link, but it is worth keeping in mind that few if any joints have been observed to fail because of this change mechanism.
Of the high-tin alloys evaluated, tin silver has the longest history, having been used on early projects during the mid 1990s. It has good mechanical properties, and indeed thermal fatigue testing has shown improvements over tin-lead alloys. A eutectic, the Sn3.5Ag alloy5 melts at 221°C. Nevertheless, apart from some conservative military users, the emphasis amongst users has swung round to ternary alloys including copper, which offer reduced melting temperature and arguably better characteristics.
The tin-silver-copper (SnAgCu, hence SAC) ternary eutectic was developed as an improvement over the basic tin-silver alloy. The melting temperature is rather lower; a number of different compositions had been claimed to be eutectic or near eutectic in the region 217°C–219°C, whilst work by Handwerker and her colleagues has since shown that the true eutectic is Sn3.5Ag0.9Cu, which melts at 217°C.
The shaded area in Figure 2 shows that a range of different compositions near the eutectic have a liquidus temperature that is under 10°C higher than the eutectic temperature.
Of course a phase diagram shows much more than the compositions and temperatures at which materials are solid, liquid, or a mixture of the two phases; it is also an indication of when different solid phases of the same material are stable (like the transition to α-tin that takes place at −13°C in the absence of any inhibitors), and also shows the limit of solid solubility of one element in another at a given temperature – recall the description given of the solidification process of eutectic tin-lead solder, and the formation of lamellae containing alternately lead-rich and tin-rich phases.
The situation is considerable more complicated when there are a number of solid phases, each with a different crystal structure, as is the case with other binary (2-element) systems, but at least the phase diagram look comparable. What becomes much more difficult is to show the same information when the alloy is of more than two constituents. If we look at the phase diagram of SAC, for example, we find that it looks something like Figure 2 above.
We are not trying to turn you into metallurgists, and this area is extremely complex. If you want to follow it further, then there is additional explanation of the meaning of phase diagrams at this link.
The Sn3.8Ag0.7Cu alloy was recommended for general purpose use by the (formative, but relatively early) Brite-Euram project as performing better in terms of reliability and solderability than tin-silver and tin-copper. Brite-Euram researchers also recommended the addition of antimony (0.5%), particularly for wave soldering, as this strengthens the alloy and further increases reliability.
In recent years there has been considerable activity in this area during the lead up to 1 July 2006. Much of the focus has been on solders with lower silver content, and a promising contender is SAC305, interpreted as Sn3.0Ag0.5Cu. It is this alloy that came out top in the collaborative research carried out by IPC during 2004/05, but some users prefer the earlier SAC387, and there is considerable interest in alloys that contain as little as 0.5% silver.
The cost of the alloys have particular consideration for wave soldering, where the bulk of the cost is in the raw material, rather than the conversion of raw material into a paste. Adding a percent or two of silver can make a significant difference to the overall cost of buying the 700kg of material needed for a typical large wave-soldering machine!
There is, however, a tin-copper eutectic (Sn0.7Cu) that melts at 227ºC and is fairly close in price to tin-lead. The conclusions from the IDEALS project were that “the melting point of SnCu is too high for wave soldering at acceptable temperature levels”, and Biglari and Oddy recommended using eutectic Sn3.8Ag0.7Cu0.25Sb (SACS) for general-purpose wave soldering, and non-eutectic Sn4Bi1Ag2Sb (SBAS).
Whilst the unmodified eutectic has been criticised, a nickel-stabilised Sn0.7Cu alloy has been patented by Nihon Superior, and is claimed to give substantially improved results. For example, it is possible to achieve bridge-free wave soldering with a solder bath temperature of around 255°C, which is within the range that paper-phenolic can handle for the 3–4 seconds it takes for any part of the board to pass through chip and laminar waves. It has been claimed that Matsushita have produced several million VCRs with a paper-phenolic board wave-soldered with this alloy, which has also been used in both horizontal and vertical systems to produce a lead-free HASL finish.
A comparison of tin-silver-copper lead-free solder alloys
by Karl Seelig and David Suraski
Lead-free Wave Solder Alloy Selection: Reliability Is Key
by LeRoy Boone et al: search at SMT magazine under Lead-free wave solder alloy selection
It is important to remember that the formulation of the alloy used in the solder is not the same as the composition of the joint, because the solder reacts with the joint materials. Typically at least some copper from the board foil will dissolve, and any fusible surface finish will disperse in the joint to a greater or lesser extent. So the joint is both contaminated and potentially not homogenous.
A further complication is added if the product is wave-soldered, because the bulk solder in the pot tends to pick up contaminants, dissolved from the board passing through, and also changes in composition as dross is removed. Tin being more reactive than other solder constituents, the dross tends to contain higher levels of tin oxide; when the bath is replenished, the composition of the solder added may need to be different from the initial charge in order to restore the intended composition. For example, with a tin-lead bath, additions of pure tin need to be made from time to time, rather than simply adding more eutectic solder. If the tin concentration is allowed to reduce, this produces an alloy with an extended pasty range and inferior performance.
The pick-up of copper in the solder has an even more serious effect than tin depletion. Common practice with tin-lead alloys has been to replace solder when the percentage of copper increased to around 0.3%, at which level the joint appearance began to deteriorate; with Sn0.7Cu, adding the same extra copper has a more substantial effect in increasing the melting temperature, as shown by the ‘copper curve’ in Figure 3.
The melting temperature of the SAC alloy, at around 217°C is still regarded by many users as being too high, particularly if there are sensitive components. Can one add anything to the system to depress the melting temperature? After all, we know that adding silver reduced the melting point of tin, and adding copper further depresses the melting temperature. Bismuth is an attractive candidate, despite reservations on environmental grounds, and a number of SAC + bismuth alloys have been patented. Jenny Hwang suggests that the optimal composition is 93.3 Sn, 3.1Ag, 3.1 Bi, 0.5Cu, as this has the finest microstructure. Its melting range is 209°C to 212°C, so it has both a small pasty range and a liquidus temperature 5/9° lower than SAC and SA eutectics.
Bismuth is not the only element that can be added to depress the melting point of SAC solder. For example, every percent of indium up to around 12% suppresses the melting temperature by about 1.8°C. As examples, Sn4.1Ag0.5Cu8In melts at 195°–201°C, and Sn4.1Ag0.5Cu12In melts at 185–195°C. As with the bismuth alloys, there are patent implications.
An advantage of adding indium is to enhance the fatigue life, and Sn3Ag0.5Cu8In is described as offering the best balance of properties with high fatigue resistance. Its melting temperatures (196–202°C) have a narrow pasty range and acceptable wetting characteristics. But of course the material is expensive. There is also some concern that, with a high indium content, there is the possibility for the growth of a tin-indium binary eutectic that melts at 117°C, a similar problem to that afflicting bismuth alloys.
Given that silver, bismuth and antimony have all been slated for adverse health hazards, are there any systems that contain none of these elements? Jenny Hwang promotes combinations of tin, copper, gallium and indium. Metallurgical reactions between the minor elements and the tin are considered to be key in determining the melting temperature and the way that the silver solidifies, and this latter determines the mechanical properties.
“The 13-year steady, sustained study [reported in her book, Environmentally-friendly Electronics: Lead- free Technology] of lead-free solder alloys has revealed that the ternary alloys cannot reach below 215ºC, and only quaternary alloys can reach below 215ºC.”
Surface Mount Technology, November 2003
Specific compositions (all patented) contain 92.8–93.0% tin, 0.5%–0.7% copper, in combination with the optimal 6% indium and 0.5% of gallium to enhance fatigue resistance. All these compositions have a narrow pasty range (5°C) and solidus temperatures around 209°C. the mechanical performance, in terms of fatigue resistance, is impressive being stronger and having an extended fatigue life. However, one has to be careful about which comparisons are being made.
Ignoring concerns about reliability, which may be more or less true, the major reason affecting the uptake of these more complex alloys is actually the substantially higher cost of the raw materials, and the patent situation surrounding them. Nevertheless, attempts to make a ‘next generation’ alloy continue to absorb substantial coverage in the media. Perhaps more promising in the longer term will be the work being carried out by EFSOT (Next Generation Environment-Friendly Soldering Technologies), an IMS project with participation from Japan, Korea and the European Union, which is believed to be focusing on improvements to the way in which the materials are formulated and combined, and to their resulting microstructure, rather than on changing the basic alloys themselves.
High-melting alloys are important within the electronics industry, both for die attach and solder balls and columns for BGA devices; in the former case, solder is used instead of silver-filled epoxy. It has thermal and electrical conductivity an order of magnitude higher; in the latter case to prevent the collapse of the ball/column.
The most commonly-used alloys have been lead-based with the addition of 10% silver (melting point 302°C) or 5% tin (melting point 320°C). The only lead-free material with any track record is the so called ‘J alloy’ (Sn25Ag10Sb). This has a melting point of 365°C, but is not ductile and therefore prone to thermal fatigue failure. J alloy was introduced as a cheaper alternative to eutectic bonding of a silicon die to a gold-plated header, which produces a silicon-gold eutectic melting at 379°C that acts as a solder, but requires expensive heavy gold plating. Although using sophisticated techniques of manufacture to ensure a fine grain structure, the results from this alloy are disappointing.
Lack of any real solution for high-temperature solders are closely intertwined with ongoing discussions about exemptions to the RoHS Directive for military and other critical applications, and to the current exemption of power devices.
[ back to top ]
IPC defines a flux as “a chemically and physically active formula which promotes wetting of a metal surface by molten solder, by removing the oxide or other surface films from the base metals and the solder. The flux also protects the surfaces from re-oxidation during soldering and alters the surface tension of the molten solder and the base metal.”
Whilst soldering can take place by the application of heat alone, the joints that result are generally poor in visual quality and unreliable in performance, even assuming than one can persuade the parts to wet at all. So in any consideration of soldering we have to find some means of removing the surface films from the parts to be soldered and protect the surfaces from re-oxidation. The material that we use for this is generally referred to as a ‘flux’, referring to the fact that it helps the materials to flow.
So, what is a flux? And what is a flux made of? There are some answers to this in our paper Fluxes for soldering, as well as a description of the different types of flux and how they are classified. This is worth at least skim-reading, so that you understand some of terminology relating to the type of flux.
At the very minimum, you should understand the different types of flux, and the impact that the choice of flux makes on the quality of the joint (particularly with hard-to-solder surfaces), on the nature and quantity of residues left after soldering, and on whether the assembly needs to be cleaned after soldering, or the residues may be left on the circuit with no adverse impact on life performance.
One element that is only partially addressed by our paper refers to the chemical nature of the flux, in particular the flux carrier for fluxes used as liquids for wave or hand soldering. Traditionally fluxes had a base of volatile organic compounds (VOCs), formulated to give the correct flow characteristics without so much solvent loss during operation that the characteristics changed significantly. [Early materials, using alcohols, had proven difficult to use, and much more sophisticated materials were developed] However, all such organic materials have the potential to evaporate, and many of them have been implicated in the generation of photochemical smog, and also have a greenhouse gas effect. A consequence of the concentration on VOC management has been the introduction of increasing numbers of fluxes which have a water base. This is one case where it is important to read the label, since some water-based fluxes may also have a VOC content! More about VOCs and their ability to harm the environment at this link.
Apart from the range of fluxes that have colophony or equivalent as their primary fluxing ingredient (aided by activators), there are other water-based preparation fluids that are extremely effective, but require careful cleaning after assembly. Such fluxes come with the cost of extra processing, but are very useful if surfaces are badly oxidised or otherwise hard to solder.
A final consideration on fluxes is whether the materials used at different soldering stages are compatible with each other. For example, fluxes are used in solder paste, at wave soldering, and during hand soldering and repair operations, and it is not unknown for the last of these to be incompatible with either or both of the first two. The result is that attempts made to clean after hand soldering and/or repair lead to significant contamination. It is instructive to search under "white deposit" on one of the technical email chat sites, such as those run by IPC and SMART Group.
[ back to top ]
Solder paste is a “homogenous and kinetically stable mixture of solder alloy powder, flux and vehicle, which is capable of forming metallurgical bonds at a given set of soldering conditions and can be readily adapted to automated production in making reliable and consistent solder joints”.
Perhaps Jennie Hwang’s description of solder paste is less than friendly, but it is a useful indication that we are looking for a consistent material, which needs to be homogenous, but tends to segregate because of the high density of the solder particles, and is a mixture with three elements. However, in practice, pastes are made of two components, solder powder and a carrying medium that is usually referred to as a “flux vehicle”. This combines the active principle of a flux (assisting soldering) with a complex set of additives that give the solder paste the desired flow characteristics. Of course, the full truth is more complex, and we invite you to read three papers that explain as much as you will need to know about this material.
The first of these, Solder paste basics, describes the constituents of solder paste in more detail, leading on to the second, Solder paste characteristics, which gives an insight into the flow characteristics of the material, and how these are controlled. It also touches on the use made of the solder paste as a temporary adhesive during assembly, and a common problem of solder balling, which occurs at reflow. To top it off, there is an explanation of what the “metal content” quoted on your paste pot actually means, and the impact that this has on the rheology of the paste.
Rheology is a term that you may not have come across, and there are other comparable terms hidden within these two papers. Depending on your memory of basic physics, you may prefer to make a start by reading our paper on Viscosity and flow.
The flow behaviour directly affects the way in which the paste is deposited, and therefore:
Each method needs different paste flow characteristics and correspondingly different formulations. Table 1 gives a set of representative specifications for metal content and viscosity which is occasionally still seen in the literature, although most pastes are squarely targeted at the stencil printing market. Remember, however, that these figures are very much simplified and that, for example, satisfactory pastes for stencil printing can be formulated with a wide range of metal contents.
|nominal metal content (%)||range of viscosity (Pa.s)||application method|
Whilst the processes nowadays have only specialist applications, it is worth noting that the changes needed to fit the application affect more than just the viscosity of the material:
Note that rheology is also important even after the paste has been printed or dispensed; being a fluid, it tends to slump and spread. This spreading may cause solder bridging between the pads and inadequate solder joint stand-off height. The degree of ‘cold slump’ depends on:
Some cold slump may also be seen with dried pastes which are left for an extended period, in the form of ‘crumbling’ at the edges, caused by loss of solvent. A second type of slump – ‘hot slump’ – is also generally recognised. This occurs during the preheat part of the reflow cycle, where the flux vehicle is more mobile and may be less able to keep the heavy solder particles in suspension, so the paste tends to spread or slump due to the effect of gravity.
In other words, the properties of the material used have an effect throughout the assembly process.
The key fact to remember about solder paste is that manufacturing solder balls of consistent size and oxide free is a serious technology challenge, and formulating an acceptable flux vehicle demands experience in the application of science. Whilst it is not difficult to persuade a paste to print properly, since the advent of lead-free soldering and its higher temperatures it has become more difficult to create a flux that will be effective in promoting wetting and flow without leaving harmful degradation products.
This is why our third paper on Practical solder paste issues states “Designing a reliable product is complex, and some of the performance parameters are trade-offs . . . every ingredient plays a role in the final performance of the paste and, most of the time, all the ingredients interact in determining paste behaviour. Be aware that materials which may appear similar may have totally different characteristics in the application!”
You probably won’t want to read that document at this stage, but it contains information on how to identify the paste you receive, and how the paste will have been tested by the manufacturer and can be tested by your laboratory. Here there are some practical demonstrations, and not just laboratory tests. And there is a final section on paste management, a reminder of both health and safety aspects of using pastes in the work place, and some recommendations on its storage and preparation for use. It is not unusual for paste to be stored in a refrigerator, in order to extend its shelf life, but putting cold paste on a stencil is not a good idea.
Thank you for staying with us to the end of this unit. We hope that you have at least identified the sources of information to which you can return later during the module. You could bookmark this page, or alternatively locate the topics on the module map.
[ back to top ]