September - October 1999 - ChipScale Review

September - October 1999


eMail the Editor

An Expert Looks at the Issues™

Dennis Bernier

Dennis Bernier is vice president, research & development for the Kester Solder Division of Litton Industries, Des Plaines, Ill.

During his 28 years with Kester, he has been responsible for product development and all laboratory testing operations.

Prior to joining Kester, Mr. Bernier was product manager and technical director for 11 years at Marshall Industries, a distributor of soldering materials and electronic components.

He received his bachelor's degree in chemistry from California State University and a master's degree in project management from the Keller Graduate School of Management.

QWhat major changes should we look for in solder over the next few years?

ABesides using the right amount of heat and having solderable surfaces, the soldering process involves the selection and use of flux and solder. Limiting this reply to solder, we can expect to see greater use of lead-free solder alloys, increased use of fine-diameter wire solder and improvements in solder paste.

Solder alloys without lead, most commonly tin alloyed with silver, copper, antimony, bismuth or indium, have existed for decades. These same solders, tin mixed with one or more of the other metals, are the lead-free combinations of choice for the electronics industry.

As electronic interconnections become smaller, there will be greater need for high-quality, fine-diameter (0.25-0.38 mm) solder wire with flux cores that are compatible with the other fluxes used in the soldering process.

We also expect to see next-generation solder paste formulations that incorporate solder particles of 5-15 microns/diameter for fine-pitch soldering and die bumping applications.

QWhat's the status of lead-free solder development?

ALead-free solder alloys have been used for many years in the electronics industry for applications that require higher-melting solder than the tin-lead alloys normally used for soldering.

The flurry of activity for lead-free solder alloys started with the U.S.A. Safe Drinking Water Act of 1986. Pressure to eliminate the use of lead has since been on electronics, an industry that incorporates into its products less than 1 percent of the lead used in the world.

Legislation in Europe and Japan has established deadlines to reduce or eliminate lead in electronics over the next few years. Some companies are already using a limited amount of tin-silver, tin-copper, and tin-silver-copper solder alloys.

The concern right now is with the ability to use solder alloys with melting temperatures and strengths that are higher than the conventional tin-lead solders that were used as the electronics industry developed.

QWhat advances have been made in solder substitutes for interconnections?

AAt this point, there are no real advances in developing substitutes for solder. The use of solder as a material for joining components to substrates is very reliable and cost-effective. Improvements have been made in both isotropic and anisotropic epoxy formulations, but very little is being used as a direct substitute for solder.

Conductive epoxies are a similar technology to what was used 30 years ago. Components that are bonded with epoxy to a substrate cannot be easily replaced.

Even though epoxies are used as reflow encapsulants to flux and then cured to help take up stresses, solder is still making the electrical connections for ball grid arrays and flip chips.

Electro-optical technology is on the horizon. Eventually there will be the potential to replace solder entirely. Imagine a circuit board with fibers instead of copper traces. Components could be bonded to the board with optical-quality polymer, using no solder, no flux, no metals, no electrical interference and almost no heat.

QWhat are the significant issues in using solder for wafer bumping?

AThere are several ways to apply the solder to the wafer. Electroplating and evaporation techniques are being used. There are some specific issues with the use of solder paste to accomplish the bumping.

One issue is the size of the bump. Solder paste is readily available with solder particles with diameters of 25-45 microns (Type 3) and 20-38 microns (Type 4) and all sorts of mixtures of these.

The conventional solder paste formulations have only limited use when trying to solder-bump a wafer terminal that is only 50 microns across. Therefore, a key issue will be the ability to produce good quality solder powder of Type 6 (5-15 microns diameter) without excessive finer particles. This is beyond the limit of currently used particle screening capability.

Another significant issue with solder paste is the flux portion, which is about 10 percent of the formula by weight. The flux should not bleed out around the solder paste deposit and should allow for easy removal.

QWhat are the key issues that users of solderfor electronics often overlook?

AOne key issue is the effect of contamination. The contamination can be on the metal surfaces that are going to be soldered, or impurities can be in the solder or in the flux.

Melted solder will only make an acceptable bond to clean metal surfaces. For soldering reliability, we always like to see components with leads hot-solder dipped, so there are no exposed, difficult-to-solder metals such as nickel-iron.

It might seem like a good idea to coat a component lead with tin by electroplating or electroless deposition. However, if the metal under the tin coating is not clean enough to solder, when the tin melts off into the melted solder, the soldering reliability is lost.

For wave soldering, impurities in the solder, such as copper, cadmium, zinc, aluminum and gold, will affect the quality of the soldering. Contaminated solder, even though it may contain lead, can be recycled and is not considered hazardous for shipping to a recycler.

Contamination in the soldering flux, such as oil or water from air lines or impurities from the circuit board assembly, can affect the quality of the soldering.

QHow is acceptance of "no-clean" soldering?

AThere is generally wide acceptance of the mild, no-clean fluxes used for wave soldering, hand soldering and reflow soldering with solder paste. The cost savings experienced by not having to remove the flux residues can be enormous. Of course, the electronics assembly must be able to operate with the flux residue remaining. This requires the use of a very mild flux that can pass electrical resistivity testing at high temperature and humidity.

Complete removal of the flux residue after soldering is required for some applications, such as military equipment, medical electronics and some high-voltage applications. Some no-clean fluxes can be removed effectively and some cannot.

QHow does the growing acceptance of flip chips and CSPs affect the quality demands placed on solder?

AThe use of flip chips and chip-scale packages puts everything on a smaller, and therefore more critical, scale. The solder used now for bumping is tin-lead eutectic or high-lead content balls or columns. The technology for manufacturing spheres, solder powder and fine wire is just barely keeping up with the requirements.

What will happen if environmental legislation suddenly requires lead-free electronics? Can the lead-free substitute alloys be supplied at the same quality level as the tin-lead alloys? Will the lead-free solder paste formulations perform the same at the higher temperatures for reflow? Can the electronic components withstand the melting points of the lead-free alloys?

All of these questions must be addressed before the electronics industry can make any changes in the solder alloy composition used for assembly bonding.



Chip Scale Review o 7291 Coronado Drive, Suite 8 o San Jose, CA 95129 o Email: editor@chipscalereview.com



Dennis Bernier on Solder, 99/10/11, 99/10/12, ID=9910/featuree1
Keywords=bm00

© 1999 ChipScale REVIEW