We talk about ‘solder paste’ or ‘solder cream’ without necessarily having a clear idea of the range of materials involved and their properties. Let’s review the requirements for this material.
The paste must have sufficient fluxing action to enable that solder mass to wet the surfaces to be joined. This combination of properties is produced by controlling, firstly the materials from which it is made and secondly the flow properties of the paste.
The original solder paints were blown metal powder mixed with thickened zinc chloride flux base. They replaced an earlier tinning method, where the operator applied a tinman’s stick and blowtorch, then wiped the area with a moleskin to remove the surplus!
Early pastes for electronic assembly applications, intended for dispensing or screen printing, were likewise comparatively crude, but the technology has developed considerably over the past 20 years, and there is an ever-increasing rate of innovation, fuelled by the increasing importance of surface mount technology.
Changes in solder pastes have come about by progressive development and optimisation, aimed at:
Many of these aspects are interrelated.
Around 1990 there were two major influences for change:
[ back to top ]
Solder paste is a suspension of solder particles in a flux-containing printing vehicle (Figure 1), in which the shape and size of the particles and the flow properties (‘rheology’) of the flux vehicle are matched to the method of paste application most appropriate to the design of SM assembly.
Remember that the solder alloy powder is the only permanent part of the metallurgical bond. At the end of the soldering cycle, all the printing vehicle and flux components are present only as residues, having been either volatilised during reflow, decomposed, reacted chemically, or removed during a subsequent cleaning step. However, all are crucial to forming reliable joints.
The flux vehicle contains both flux and other substances that give the solder cream its ‘pasty’ nature. Its tasks are to:
The term ‘flux’ is used fairly loosely to describe a wide range of materials which have a fluxing action. These include thin, solvent-rich liquid fluxes and the highly viscous materials used in formulating pastes, as well as referring to solid flux bases such as rosin. To reduce the possibility of misunderstanding, we have tried to be consistent in using the term flux base for this last group of materials, and in referring to the mixture of flux, activators and printing vehicle used in solder paste manufacture as flux vehicle.
The flux vehicle generally has four constituents:
The flux, solvent and additives play a major role in determining the viscosity of the solder paste and the processing time prior to placement. Additives give the paste the correct flow characteristics.
Two types of chemical are involved in the fluxing function: the flux base and the activator, which enhances the fluxing ability of the system. Until recently, rosin (or colophony) has been the most widely-used and reliable flux base, presented in several standard strengths:
Other pastes use the same range of alternative formulations (based on synthetic resins, or water-based and intended for aqueous cleaning).
Remember always that different manufacturers make totally different pastes, even with the same nominal specification, and we can never assume that claimed replacements are either exactly equivalent or even fit for purpose.
Solder paste is a suspension of solder particles in a flux vehicle. The previous section has concentrated on performance aspects of the flux; this section deals with the metal which forms the major proportion of the paste by weight, even if only around half by volume.
Solder paste can be made from any solder alloy, but the most frequently used materials are tin-silver copper alloys, replacing the earlier tin-lead and tin-lead-silver eutectics.
Key parameters to be controlled are:
These are tailored to fit the needs of applications with respect to rheology and solder paste deposition and land pattern.
In the past, some manufacturers advocated the use of non-spherical particles which could ‘interlock’. There was some evidence that using spherical particles alone could induce slump and outflow of the paste as the flux melted.
Slump is when the particles sink just before melting occurs, leading either to solder bridging or unequal deposits (where paste on adjacent tracks was previously touching and then split unequally back onto the pads).
Flow is when particles which spread apart as the flux melts are ‘stranded’ at the edges of the fluxed area and remain as undesirable discrete solder balls, not drawn into the main solder fillet.
For some applications, and a small number of users, a proportion of non-spherical solder particles may still be specified, in order to retain as good a shape profile as possible throughout reflow. However, the prevailing belief in the industry is that solder particles should be as nearly spherical as possible (Figure 2), for the following reasons:
Source: DEK Printing Machines
Solder particles are sorted into various size ranges by sieving
The relative dimensions of the meshes used are shown in Table 1.
|Mesh size||Particle sizes|
|-325||less than 45µm diameter|
|-270||less than 55µm diameter|
|-200||less than 75µm diameter|
|-200, +325||between 45µm and 75µm diameter|
The mesh size is in units of ‘lines-per-inch’, with
the minus sign meaning the particle fraction that has passed through
the mesh is used and the plus sign meaning the fraction that has
not passed through is used.
The choice of particle size is a compromise between ease of application and definition of the applied paste pattern as against the problems of surface oxide and solder balling. The finer the particles:
A solder paste can, in principle, be manufactured using any size range or combination of size ranges specified by the customer, but solder paste manufacturers have, in the main, standardised on a limited number of particle size distributions. As shown in Table 2 below, the American ANSI/J-STD-005 standard, which has superseded IPC specification IPC-SF-819, defines six sizes of ball.
|Type||<0.005%larger than||<1%larger than||>90%between||>10%smaller than|
‘Dust’, and the similarly-sized ‘satellites’ which sometimes accompany solder balls, are avoided because of their high oxide content. It is also important to avoid either over-size balls or aggregates of balls which will not pass through the stencil apertures. Both these requirements can be seen in the ANSI/J-STD-005 specification, which includes a specific test for maximum particle size.
There is obviously a distribution of particle shape and size, depending on the method of manufacture, and yields frequently do not exceed 50% – the smaller the ball and the tighter the distribution, the lower the yield, which explains why fine pitch pastes are more expensive. To produce good quality powders with a reasonable yield has been the major thrust of work in recent years, and making powder to very accurate size distributions is ‘the trade secret’.
The oxide thickness on solder will depend on its exposure to the air, but some permanent oxide film will be established on a clean surface unless the oxide level is below 15ppm. Powder will start growing an oxide film immediately after manufacture, although the rate will depend on such factors as temperature and the presence of flux.
Unfortunately, the ratio of surface area to volume of a sphere increases as the particle size reduces and so, for a given oxide thickness, more oxide will be present in the paste. The degree to which this happens is shown in Figure 3.
after Lea 1988
Truly oxide free powder is hardly attainable, and although excessive oxide content is detrimental to solderability and can cause solder balling, powder containing oxygen in the range 50–200ppm is generally considered satisfactory.
A shot tower method (as for lead shot) is still in use, but only for the large balls used for ball grid arrays (BGAs). Solder powders are produced by a variety of atomisation processes, where molten alloy is broken into particles using a jet of gas, a spinning disc, or an ‘ultrasonic horn’. The powders produced have relatively high density, a good flow rate, and are spherical in shape.
Figure 4 is a schematic of an inert gas atomisation system, consisting of a vacuum induction furnace, tundish, ring nozzle, atomisation tower, cyclone, and powder collection container. The solder is melted under inert gas to avoid evaporating the alloy components, and then dosed into the atomisation tower where the melt is disintegrated into powder by an energy-rich stream of inert gas. The process is carried out under inert gas at atmospheric pressure, and is able to produce high-quality powder.
The shape of solder particles depends critically on the powder production process. Non-spherical particles and agglomerated clumps must be eliminated at atomisation, since this cannot be done during the next stage of sieving in nitrogen. Whilst this process will separate powders of different sizes, irregularly shaped particles can pass a sieve with one dimension much longer than the mesh size.
For reasons associated with the nature of the process of blowing powder, its conversion into paste is often carried out at a different location. Care has to be taken during transit and storage to avoid both contamination by chemicals or foreign bodies and oxidation. Powder is stored in nitrogen, and there is a valid argument for enclosing the entire process, or using clean rooms, to ensure suitable conditions for manufacture
Why is it important that the solder particles used in paste are:
a) spherical in shape?
b) of consistent size?
c) free from oxide?
[ back to top ]
Author: Martin Tarr
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 2.0 Licence.
Terms and conditions apply.