Solder Grain Structure And Cooling Rate
A solder joint does not freeze at a single temperature. The alloy solidifies over a range, during which crystals grow from the cooling surface into the liquid, and the size and shape of the resulting grains depend on how quickly the joint was cooled. That rate is set by the profile, by the thermal mass of the assembly, and by the geometry of the joint, and it decides the mechanical properties of the joint that will carry the product for the rest of its life.
This article explains what happens as a joint solidifies, how the cooling rate changes the structure, what the intermetallic layer does, and how the cooling is controlled and verified.
What Happens On Solidification
As the joint cools below the liquidus, crystals of the primary phase begin to form and grow into the remaining liquid as branched structures called dendrites. A lead free tin silver copper alloy forms a tin rich primary phase with intermetallic particles of silver and copper dispersed in it, and the size and distribution of those particles are set by the cooling rate. A fast cool leaves little time for the particles to grow, so the structure is fine; a slow cool allows them to coarsen into large plates and needles.
Supercooling is a complication that is specific to lead free alloys. The liquid can remain liquid below its nominal solidus until a nucleus forms, and the delay is several degrees or even more, which means that a thermocouple attached to the joint may read a temperature below the nominal melting point while the solder is still entirely molten. The practical consequence is that time above liquidus is measured from the nominal value in the profile but the joint may actually solidify later than the calculation suggests.
<img src="https://www.gopcba.com/wp-content/uploads/2025/08/32.png" alt="Microsection showing fine solder grain structure” />
The Effects Of Cooling Rate
A fast cooling, in the region of a few degrees per second or more, produces a fine structure with a thin intermetallic layer at the pad interface. The fine structure has better resistance to mechanical shock, because the cracks that start in service have to travel around many small particles rather than through a few large plates, and it also produces a smoother and more reflective surface. A slow cooling, on the other hand, produces a coarse structure and a thicker intermetallic, and it is associated with a dull and grainy appearance that is often mistaken for a cold joint.
Very fast cooling has its own risks. A quench can leave residual stress in the joint and in the surrounding laminate, and on a large area array package it can crack the solder at the corner balls, where the thermal gradient across the package is largest. The thermal mass of the assembly spreads the range in practice: the joints on a small, lightly populated board cool faster than those on a heavy backplane, so the same profile produces different structures on different products.
The Intermetallic Layer
The intermetallic layer between the solder and the copper pad is not a defect. It is the evidence that a metallurgical bond has formed, and it is what makes the joint conduct and hold. It is also brittle, and a layer that is too thick becomes the weakest part of the joint, because a crack can travel along it. The layer grows during soldering and continues to grow slowly in service, and its thickness at the end of the process is set by the time above liquidus and by the peak temperature, which is one of the reasons both figures are limited in the profile.
Cooling rate has a second order effect on the layer as well, because a fast cool limits the additional growth that happens while the joint is between the liquidus and the solidus. The combination that produces the best joint is therefore a profile that reaches a sufficient peak, spends the right amount of time above liquidus to form a proper fillet, and then cools quickly enough to freeze the structure while avoiding thermal shock.
Cooling And The Defects It Causes
A joint that is disturbed while it is still in the pasty range is the classic cooling defect. Movement of the board on the conveyor, vibration from a fan, or the differential expansion of a heavy component can all displace the partially solid solder, which produces a fillet with a rough, torn appearance and a structure that is mechanically weak. The window in which this can happen is shortest for an alloy with a narrow pasty range, which is one reason the alloy choice interacts with the mechanical handling of the line.
Shrinkage voids are the second consequence. A joint that solidifies slowly can develop a void where the last liquid contracts, and on a large joint that void may be significant enough to affect the thermal path. A very fast cool avoids the shrinkage but introduces the stress described earlier, so the practical answer is a controlled cooling ramp that is fast enough to refine the structure and slow enough to avoid a thermal gradient across the package.
Practical Control And Verification
The cooling section of the profile is set by the last oven zone, by the conveyor speed, and by the cooling fans, and it is often the least examined part of the profile because the peak is what everyone watches. The rate is measured with a thermocouple on the joint, and a rate below about one degree per second is slow enough to produce a coarse structure, while a rate above about four degrees per second is fast enough to raise the thermal stress on a large package.
The result is verified on a first article or a qualification build. A cross section shows the grain structure and the intermetallic thickness, and it is compared with a reference that was taken from a joint known to be acceptable. The comparison is not a substitute for a mechanical test, but it is a direct view of what the profile did, and it is the fastest way to confirm that a change in the oven or in the assembly has not changed the joint structure. The alloy context is described under lead free and leaded soldering, the disturbance caused by movement under component shift during reflow, and the acceptance framework under PCB design quality characteristics.
Process Control and Verification
On a design of this kind, microsection is the item that decides how the rest of the board is arranged. A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance. 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.
FAQ
Is a dull joint a cold joint? Not necessarily. A lead free alloy cooled slowly, or one with a coarse structure, has a dull and grainy appearance while being perfectly sound. The judgement has to be based on the fillet shape and on the evidence rather than on the surface finish alone.
Why does the thermocouple read below the melting point while the solder is liquid? Because lead free alloys supercool. The liquid can persist several degrees below its nominal solidus until a nucleus forms, so the phase change happens later than the profile suggests.
How is the cooling rate measured? From the same thermocouple that records the profile, by taking the slope of the curve between the peak and the solidus. A rate between about two and four degrees per second is a common target for a lead free assembly.
