Weight change as a function of time and temperature aids reflow profiling.

One of the inconveniences of flux residue, besides poor appearance, is its deleterious effect on in-circuit testing. During testing, test pins might fail to penetrate the hard residue that can be located on the test points, resulting in misclassification of nondefective boards. Therefore, flux residues should be minimized.

Another important issue is paste compatibility with the reflow process. Depending on the type of reflow profile – e.g., linear or ramp soak spike profiles – and the brand, solder pastes behave differently. Critical areas on the reflow profile to ensure good solder joint formation are at the end of the soak and at the peak where reflow temperatures are at their highest, resulting in higher paste and surface finish oxidation. In addition, results from the thermal analysis allow us to design flux management systems that will improve reflow oven performance: Where to exhaust air or nitrogen in the oven, and how to filter vapors, are critical issues to consider during oven and maintenance design.

To understand paste compatibility with reflow profiles, possible effects on pin probe testability, and the required maintenance, thermal analysis of the new Pb-free solder pastes can be used. A thermal gravimetric analysis (TGA) instrument is useful in understanding and comparing behavior of different Pb-free solder pastes. TGA measures the weight change as a function of time and temperature. In this experiment, a clean copper coupon is manually printed with solder paste using a 0.005" thick mini-stencil. As a result, the amount of solder paste deposited is in the range of five to 10 mg. The coupon is placed in a ceramic holder on the TGA and subjected to a 4-min. Pb-free ramp/soak/spike under a nitrogen environment.

The amount of flux evaporation was calculated by oven zones (Figure 1). In this case, an oven with 10 heating and three cooling zones was simulated. Normally, flux starts evaporating at 120°C, at which only 1% of the flux evaporates in the preheat areas. During soak, flux evaporation depends on the chemistry and is usually between 20 and 30%. In this evaluation, extreme values of 45% were observed, which result in poor performance of solder paste in the peak areas. On the peak area, flux loses typically 45 to 60% of its weight. Of 10 solder pastes tested, 10 to 20% of the flux still evaporates on the cooling zones, resulting in potential condensation of the vapors and contamination of the oven’s internal parts.


In addition to the previous test, 3- and 5-min. Pb-free profiles were performed on seven solder pastes of different alloys and suppliers, and the residue amount recorded. The results are shown in Table 1. The residue amount was calculated as % weight. For example, if after reflow, the sample had a weight of 92.6% of the initial amount and the metal content is specified as 88.5%, then the amount of flux residue is (92.6-88.5)/(100-88.5) = 35.4%.


For all 3-min. profiles, all solder pastes left between 35% and 52% of residues, whereas the amount is between 30% and 54% for 5-min. profiles. Solder pastes from Samples 1, 3 and 4 show small differences between the two profiles, indicating these pastes are compatible to reflow profile with higher conveyor speeds. Also, some solder pastes need longer reflow profiles to ensure less residue on the assembly.

The data analysis allows us to estimate the amount and where in the reflow profile flux evaporation occurs. Solder paste with poor activity in the peak areas will perform poorly and poor slump resistance can be observed, resulting in bridging or solder beading. When flux residue is a concern, proper reflow profiles that reduce the amount of residues on the assembly also can be identified. In addition, scheduled maintenance can be proposed, depending on solder paste behavior.

Denis Barbini, Ph.D., is advanced technologies manager, Vitronics Soltec (vitronics-soltec.com); dbarbini@vsww.com.