Solder Joint Reliability and Intermetallic Growth

A solder joint is not a mechanical contact between two metals; it is an alloy that forms during soldering and continues to change afterwards. Understanding that layer, and how it grows, explains why a joint that passes inspection can still fail after a thousand thermal cycles.

How the Joint Forms

When molten solder wets a copper pad, the tin in the alloy reacts with the copper and forms an intermetallic compound at the interface, a thin layer with a crystal structure different from either metal. That layer is what makes the joint electrically continuous and mechanically strong, and it forms within seconds at soldering temperature.

Wetting is the prerequisite. Without it the solder beads on the surface and no intermetallic layer forms, which is why flux is used and why an oxidised pad cannot be soldered no matter how long the iron is held against it. Our solderability notes describe how the wetting behaviour is measured.

Intermetallic Growth During Service

The intermetallic layer continues to thicken after the joint is made, because the same diffusion that formed it keeps working at any temperature above absolute zero. The rate is strongly temperature dependent, so a joint in a hot environment grows a thicker layer than one in a cool one over the same time.

A thin layer, a few micrometres, is desirable. A layer of several tens of micrometres is brittle, and a crack that initiates inside it propagates easily because there is no ductile material to arrest it. That is the mechanism behind many joints that fail after long service without any mechanical overload.

Why Thermal Cycling Breaks Joints

The materials in a joint have different coefficients of thermal expansion. The component body, the solder and the board move by different amounts as the temperature changes, and the joint has to absorb the difference. Because the joint is constrained at both ends, the strain accumulates as plastic deformation rather than as a simple elastic stretch.

The damage appears as cracks that start at the interface and propagate across the joint, and it normally begins at the corner furthest from the neutral point of the package. That is why the outermost ball on a large area array package is usually the first to fail, and why the size of the package matters more than the number of cycles.

The Effect of the Void Fraction

Voids reduce the cross-section that carries the current and the load, and they also act as stress concentrators. A small number of scattered voids has little effect on reliability, while a single large void at the interface can halve the life of the joint.

Process control therefore focuses on the largest void rather than on the average void percentage. A joint spec that sets a limit on the largest single void is more useful than one that sets a limit on the total volume. Our X-ray inspection notes describe how the measurement is made.

Standoff and Joint Geometry

The geometry of the joint determines how the strain is distributed. A joint with a large standoff, meaning the component sits further from the board, is more compliant and survives more cycling than a flat, thin joint. That is one reason a ball grid array with a tall ball outperforms a package with a very low standoff.

For chip components the equivalent variable is the fillet. A solder fillet that rises up the end termination spreads the load over a larger area and reduces the stress at the interface, while a joint that only wets the pad beneath the part concentrates it.

Contamination and Its Consequences

Contamination at the interface changes the chemistry of the intermetallic layer. Gold, for example, dissolves rapidly in molten solder and forms a brittle compound that weakens the joint, which is why gold-plated surfaces are kept thin and why the plating thickness is specified rather than left to chance.

Residue from a poorly cleaned board is a different problem: it may be harmless electrically and still promote corrosion at the joint over years of service. Where the product is exposed to humidity, the cleanliness of the assembly is a reliability parameter rather than a cosmetic one.

Testing the Reliability of a Joint

Accelerated thermal cycling is the standard method, and the result is reported as a number of cycles to failure for a defined profile. The value applies only to the profile used, so a figure quoted without the temperature range and dwell times cannot be compared with anything.

Other tests answer other questions. A shear test measures the strength of a single joint and is useful for monitoring a process, while a drop test measures the resistance of the whole assembly to mechanical shock. Our component reliability notes describe how the tests are selected.

What the Designer Controls

The designer controls the pad geometry, the thermal mass connected to the pad, the position of the component relative to the neutral point of the board, and the choice of surface finish. Each of these affects the strain the joint will see, and all of them are decided before the first board is built.

The most powerful single decision is the placement of large packages. Keeping a large area array near the centre of the board reduces the displacement it experiences during a thermal excursion, and that change alone can multiply the life of its joints. Our thermal management notes describe the related placement rules.

Process Control and Verification

Reviewing the design before the data is released is far cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end. Where a value sits close to a process limit, the drawing should say so, since the shop can then open the process window rather than working to a nominal figure that carries no tolerance.

The measurements that matter are the repeatable ones: conductor width and spacing, annular ring, finished hole size, plating thickness and surface finish are all verifiable on a coupon that travels with the panel. Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule.

Where a value sits close to a process limit, the drawing should say so, since the shop can then open the process window rather than working to a nominal figure that carries no tolerance. Running a first article through the same checks as the production panel confirms that the two agree, and that comparison is the cheapest form of process control available at prototype stage.

FAQ

Is a thicker intermetallic layer stronger? No. The layer is brittle, so strength increases up to a point and then falls as the layer thickens. The optimum is thin, which is why a short soldering time at an adequate temperature produces a better joint than a long dwell at a low one.

Does lead-free solder change the reliability? The intermetallic chemistry changes and the melting point is higher, which raises the thermal load on the assembly. A correctly profiled lead-free joint is reliable, but the process window is narrower and the consequences of exceeding it are greater.

How does gopcb verify joint reliability? We verify it through the process rather than by testing every board: the reflow profile is recorded, the void fraction is measured by X-ray on a sample, the plating thickness is confirmed on a coupon, and the thermal cycling data from the qualification run is retained for the product family.