Lead Free

The ban on the use of lead (and other heavy metals and some flame-retardants) will be effective from 1 July 2006 . This is a tight timescale for transition to the new lead-free soldering technologies, and it requires that the necessary actions are started as soon as possible. The requirements for labelling for recycling (WEEE) will come into force in August 2005, and these will be transposed into UK law by the UK government.

No your business is not exempt. There are eight categories that are specifically covered by the ROHS Directive, and they include your industry. Other categories are only “exempted” by omission (e.g. medical, defence, aerospace). The categories listed cover essentially the products used by the general public i.e. the high volume, low cost products:

a. Large household appliances
b. Small household appliances
c. IT and telecommunication equipment
d. Consumer equipment
e. Lighting equipment
f. Electrical and electronic tools
g. Toys (including leisure and sports equipment)
h. Automatic dispensers

These are only indicative categories, and the list will be reviewed and may be altered in the future as more data on the impact on the environment and human health of other hazardous materials become available. For example, a review is expected in 2005 of both medical and monitoring equipment, both of which are included in the WEEE Directive.

There is a separate Directive, which applies to the automotive industry – the EOLV (end-of-life vehicles) Directive. This Directive, which is already in force, also bans the use of lead and other hazardous substances, but it does specifically exempt lead in solder for electronics. For the automotive industry, the EOLV Directive takes precedence over the WEEE/ROHS Directives, but the situation may change since the Commission can “on a regular basis, according to technical and scientific progress, amend, add to or delete from the list of exemptions”. Currently it is generally understood that “solders for electronic circuit boards and other electrical applications” for vehicles are exempt, but lead in solders for non-electrical applications (e.g. radiators) is banned for products put on the market after July 2003. In spite of the EOLV exemption for electronics, many manufacturers have already made the transition (or intend to) to lead-free materials and soldering processes.

After July 2006, both assemblies and components that are imported to the EU must be lead-free in compliance with the ban. The onus is on the “producer” which means anyone who manufactures and sells electrical and electronic equipment under his own brand, who resells under his own brand equipment produced by other suppliers OR who imports that equipment into a Member State”. Importers of electrical and electronic equipment (including components) into the EU are therefore considered to be “producers” and hence responsible for ensuring compliance. This is true even if the ultimate destination of the equipment (manufactured in the EU) is outside the EU. The Directive applies to products and producers irrespective of the selling technique, including distance and electronic selling.

This has not yet been confirmed, but it is anticipated that at the local level this will be the responsibility of the Trading Standards Authority. The DTI plans to announce the enforcement agency in mid-2005. Producers must demonstrate compliance with the regulations by providing the enforcement authority (on request) with satisfactory evidence. The UK government has accepted that self-declaration will form the basis of any compliance regime, and that market surveillance may be required to detect non-compliant products.

1. At the present time, No. But the component manufacturers are making significant strides in making their product lines lead-free, and constant dialogue with your suppliers is essential. There are two issues that you must consider when specifying components: (i) they must be compliant with the Directives, and (ii) they must be able to withstand the process temperatures required. (NB. The relevant standard for this is IPC/JEDEC J-STD-020C).

Yes, there are a number of issues that you should consider. It is essential that you develop a plan, which includes addressing the following supply chain issues:

1. re-qualification
2. stock retention through distribution
3. re-design due to lead-free enforced obsolescence
4. audits
5. manufacturing trials
6. establishing responsibility through supply chains

No. It is your responsibility unless you contractually ask sub-contractors to cleanse the components (i.e. check against compliance), which is likely to attract a cost.

Probably. Component manufacturers may not transition all current designs into a lead-free form, and it is necessary to maintain a close dialogue with the suppliers about their future policy. You may need to source the components from alternative suppliers, or even undertake a re-design.

Yes. You should immediately transition all parts and designs to their lead-free counterparts, and undertake a new lifetime buy for these lead-free components. For existing lifetime buys you can only sell lead-containing product for the purposes of repairing EEE, or as an extension to the original contract. New customers will have to purchase components etc. in the lead-free form.

Since this depends on customer specification, industry, and compliance with the Directives, there is no simple answer. However, many assemblers may need to consider re-designs.

There are three main actions you should undertake. First, clearly specify that you require lead-free components (this means (a) RoHS compliant, and (b) process compatible with higher temperature wave/reflow). Second, work closely with your supplier not only to ensure that he understands your requirements, but also that you fully understand the nomenclature of all labels on his components. Third, plan and optimize the timing of the transition to lead-free product (i.e. the purging of lead-containing materials and the introduction of lead-free materials).

In order to demonstrate compliance with the regulations it is necessary to provide a full Bill of Materials ( BOM ), and there are two main ways in which producers can generate the BOM. First, you can obtain suppliers’ declarations that any of materials or components used in their equipment do not contain more than the permitted level of any restricted substance). Second, you can undertake analysis of materials and components to verify suppliers’ declarations or to provide evidence where no declaration is available. The criteria for analysis will depend on the quantity of product put onto the market, the relationship with suppliers, the risk of banned substances being present, and the potential environmental impact. We recommend using suppliers’ declaration – producer analysis is more suited for specific or unique cases – but in both cases relevant records should be kept for four years after the EEE product was placed on the market. The UK government accepts this approach as a way of demonstrating due diligence.

The available evidence suggests that it is unlikely there will be any reliability issues from mixing during rework. But it is sensible to rework/repair with the same alloy (if known), and it may be sensible to label boards to indicate the solder alloy used.

No, you will have to change your process to some degree. The choice is not necessarily straightforward, and will depend on the application, thermal and temperature factors, assembly technology, component thermal sensitivity, service life, volume, cost etc. However, there is an increasing consensus for using the SnAgCu family of alloys for many applications for both reflow and wave soldering. Many data are available on their performance, most research on lead-free alloys for mass soldering has focused on them, and we recommend their use. But there are certain implications mainly regarding the higher soldering temperatures required (245-260 deg C). At these higher temperatures, there may be issues of component stability, assembly equipment robustness, materials stability etc.

There are several families of alloys commercially available as lead-free solders:

• Reflow soldering: SnAgCu, SnAgCuBi, SnAg, SnAgBi, SnZnBi, SnIn (the bismuth-containing solders are more suitable for low temperature soldering)
• Wave Soldering: SnAgCu, SnCu, SnCu(Ni 0.1%), SnCuX, where X is the small fraction of elements claimed to increase performance

No suitable alternative has yet been identified for this solder, which melts at around 302OC. This lack of an alternative has been recognized by the authorities, with the result that “lead contained in high melting temperature type solders” (i.e. tin-lead solder alloys containing more than 85% lead) has been added to the list of exemptions within the WEEE/ROHS Directives. So, until further notice you can continue to use this particular lead-containing solder.

Using SnAgCu as your lead-free solder will mean that you will have a narrower process window and will have to tighten your process control. In turn, depending on the equipment you already have, you may need to purchase new equipment. For example, with the higher soldering temperatures it may be necessary to consider pre-heating for the larger components. Other points to consider are:

Reflow soldering
Convection ovens are preferred to IR ovens to provide good temperature control and adequate temperature range; nitrogen inerting may be required to widen process window

Wave soldering
Equipment may need to be modified to avoid damage at the higher soldering temperatures; solder pot corrosion may be an issue; nitrogen inerting may be required to widen process window

Hand soldering
There should be few problems, and no new equipment is necessary; operator training may be appropriate on issues associated with the higher temperature lead-free soldering.

Yes, but there will be certain practical consequences. For example, the iron tip may dissolve more quickly requiring more frequent replacement, and this will be worse with the finer tips that are being increasingly used within the industry. In addition, a separate rework/manufacturing area with separate soldering irons may be desirable for lead-free soldering to avoid cross-contamination.

In general, there should be no problem – all the major data available suggest that reliability is not really affected by the change to lead-free soldering. Indeed, the cyclic fatigue resistance at constant temperature of SnAgCu soldered joints can be better or worse than that of SnPb soldered joints, depending on levels of strain. Hence it is vital to generate reliability data specific to your product and working temperatures.
Good long-term reliability depends on appropriate choice of materials and on compatibility considerations.
The reliability of the lead-free solder joints is largely similar to that of their traditional lead-containing counterparts. So if the nature of your product means that solder joint reliability is not a major issue, then using a lead-free solder should not affect the reliability of your product. The biggest risk for SnAgCu lead free joints relative to SnPb is in a high strain environment (large devices, ceramic, directly attached with a large temperature range in service).
There may be some issues with temperature sensitive components , especially in the transitional stage. You must consider the component stability at the higher soldering temperatures e.g. with electrolytic capacitors, and with “pop-corning” in plastic encapsulated components.
Substrate reliability may be affected as a result of the higher soldering temperatures, especially when using complex fine feature pcbs.
In general, lead-free soldering technologies require increased process control to achieve the same process yields provided by traditional SnPb soldering. If attention is not paid to the narrow process window associated with lead-free soldering, then the resulting lower process yield may influence product reliability adversely.
Circuit assemblies inevitably contain a mix of components, and during thermal cycling this results in a range of strains on solder joints. If these strains are high (e.g. as a result of low stand-off, large devices, direct attachment, ceramic packages, large in-use deltaT) then it is more likely that SnPb soldered joints will be more reliable than their lead-free counterparts, and vice versa.
Where the lead concentration in the soldered joints ranges from 1-5wt%, reductions in joint strength can be encountered.

Yes, the process window is markedly narrower with lead-free soldering. A particular concern lies with the ability of the reflow ovens to provide the smaller deltaT necessary. Some manufacturers have successfully used nitrogen inerting during soldering to widen the process window. There should be no changes in process control required for printing, placement and inspection of lead-free product.

Yes, for example, maintaining the composition of the solder bath, and advice should be sought from your solder and equipment suppliers. For example, the copper levels must be taken into account when replenishing the solder bath, and copper dissolution of the tracks and pads can result in their being less than the minimum acceptable thickness. There may also be an issue with the compatibility of the solder pots, pump impellors and solder bath nozzles (or any components in contact with the solder) especially at the higher soldering temperatures associated with lead-free soldering. As a result, some manufacturers have now upgraded their products and applied protective coatings as appropriate. Process settings should be based on the results of trials or R&D exercises. Using nitrogen inerting during soldering will widen the process window. Dross rates using lead-free solders vary markedly – they can be twice that with SnPb for solders containing silver, to less than half that of SnPb. To avoid possible cross-contamination, we recommend that you use different solder recovery systems for lead-containing and lead-free soldering systems.

There is also a new so-called failure mode – fillet lifting – which sometimes occurs with leaded components in wave soldering if there is any lead in the system. The phenomenon is manifest as a lifting of the solder fillet away from the land after soldering, and is caused by differential contraction on cooling and contraction of the board in the Z-axis. It can be manifested in three types. Of these, two (“fillet lifting” and “tearing”) are not thought to cause a reliability issue (i.e. they are “benign”), and only one (“pad lifting”) is malignant. In the latter, peeling of the pad from the substrate could destroy the connection. But it is not yet clear whether or not there are any deleterious effects on either the joint or the product. Evidence suggests that the effect is minimal.

Probably, but only in minor ways associated with the appearance of the lead-free joints. For example, lead-free soldered joints are less shiny and more uneven than their traditional counterparts, and the pcb land/pad coverage tends to be lower giving rise to copper halos. However, neither appearance seems to degrade joint reliability. You may have to alter the settings of your inspection techniques, manual or automatic, to take account of the reduced contrast associated with the reduced lead content. There is a new standard, which covers the inspection procedure ( IPC-A-610D ).

If your boards already require cleaning, then you will need to continue to clean boards assembled using the new lead-free solders. But the cleaning may be more challenging, since there will probably be more corrosive residues present, and they are likely to be more tenacious and baked-on. If you already have a no-clean process, you will probably not need to clean using your new lead-free process. If you currently have to comply with a customer specification, you will have to re-qualify your new process.

The following components terminations are available: Sn, SnCu, SnBi (silver and palladium finishes are also available but may be too expensive). The following lead-free board finishes are available: tin, tin alloys, ENIG (electroplated nickel gold), silver, silver alloys, lead-free HASL (hot air solder levelling with SnAgCu or other lead-free alloy), OSP (organic solderability preservative).

There might be. If the component termination is lead-free, no compatibility issues would be expected with lead-free soldering. However, if the component termination does contain lead, the resulting lead concentration in the solder joint may be in the range 1-5wt%. If so, the there can be risks of poorer reliability. (NB. This is an area of keen research).

No. The status of exempt applications is under constant review. Moreover, in the future, as green issues mature, the combination of market push and public perception may well result in a customer’s preference for lead-free product. In addition, sourcing your BOM (Bill of Materials) may also change and you must be aware of the ramifications this may have on your processing.

Yes, but there is the potential for poor alignment and open joints. Since the usual SnPb reflow profile may not exceed the melting point of the BGAs’ lead-free solder balls, the SnPb solder paste melts but the SnAgCu solder balls do not. The lack of ball collapse may cause a lack of contact between the solder paste and the solder ball. We recommend that you use a reflow profile that ensures the SnAgCu solder balls also melt, and the lead from the molten SnPb solder paste mixes thoroughly. In doing this you should be aware of the affect on any other temperature sensitive components, however.

Yes, but they are not likely to be great and many will be of a one-off nature. Initial costs may include new equipment purchase (where necessary), existing equipment modification (for higher temperatures, inerting etc), implementing new inspection procedures, training and stock control. On-going costs may include those associated with tighter process control and increased power consumption from higher temperature soldering. The increase in materials costs will be greater for wave alloy than for solder paste, but costs may come down over time and volumes of LF materials go up. However, recycling and/or disposal costs are unknown, although there is likely to be a marketing advantage for “green” product.

There are no specific data here, but there are two aspects to be considered – component functionality, and solder joint performance. Component functionality: No additional thermal excursion is desirable since it may cause damage or de-lamination within the component, samples can be checked after tinning using scanning acoustic microscopy (SAM). For high reliability products, it is recommended that long-term component reliability is undertaken.

Solder performance: this is unlikely to be a major issue. Such data as there are suggest that if bismuth is not involved then it is probably better not to waste time and money on costly retinning. Hence try and avoid the use of bismuth-containing materials and go ahead without tinning. However, some retinning may be necessary because of “lifetime buys” of components with SnPb terminations, which will be used, when the restriction of the RoSH Directive comes into force requiring no more than 0.1 wt% lead in any homogenious material. In this case we suggest using SnAgCu (SAC) alloy as the tinning material in order to keep the tinning temperature as low as possible.

There are, as yet, no relevant long-term data available. The no-clean process has always been a balancing act between flux activation and residue activity, and for lead-free processing this balance needs to be re-established. Whilst higher process temperatures will tend to lower the flux activity (by evaporation etc), the lead-free flux formulations will probably have higher levels of activators, and the overall effects are likely to be neutral. In the absence of relevant data, it is recommended that if you want (or it is necessary) to use lead-free solder processing in the manufacture of extended life product, then SIR testing should be undertaken to assess the long-term effects of any residues.

Embrittlement is associated with the presence of gold (usually via the metallization) which results in the occurrence of gold-tin phases weakening the interface between them, and in the tin-lead eutectic. However, provided the concentration of gold in the solder does not exceed 8%, then the gold-tin compound will be present only as a minor phase and cause few problems. As yet, there are no explicit data relating to this issue, but our view is that of SnAgCu solder is used any potentially detrimental effects will be lessened, since in this case more gold can be accommodated in the solder before precipitation of the offending intermetallic.