Abstract: This paper discusses the reliability testing problem of no-clean flux, and makes a comparison and analysis based on the test. The necessity of evaluating the reliability of no-clean flux is proposed.

 

No-clean flux is a new type of flux produced with the development of the electronics industry and the needs of environmental protection. It is of great significance in solving the problem of not using CFC cleaning solvents to reduce environmental pollution and solving the cleaning difficulties caused by the assembly of fine gaps and high-density components and the compatibility issues between components and cleaning agents. Since the early 1990s, no-clean flux has been widely used in the welding production of electronic components and printed boards in the electronics industry. However, as the application range of high-density, lightweight, miniaturized, and high-performance electronic products expands, the application environment becomes increasingly complex, and the reliability requirements for products are getting higher and higher. Correspondingly, the reliability of no-clean fluxes is have put forward higher requirements. This article analyzes the test conditions of substrates after soldering using no-clean flux on the basis of experiments, and discusses the necessity of evaluating the reliability of no-clean flux.

 

The influence and microscopic mechanism of post-soldering residues of no-clean flux

 

As a no-clean flux, it must meet the following conditions: (1) The post-weld residue should be minimal; (2) The post-weld residue should remain inert and non-corrosive under temperature and humidity; (3) The post-weld residue should have a high Insulation resistance value. The so-called post-weld residue refers to the post-weld non-volatile components and residual active components in the flux, as well as the metal oxides generated by the post-weld reaction. From a physical point of view, the reaction products and residual substances may be isotropic dielectrics. The molecules of this kind of dielectric can be divided into two categories; one is non-polar molecules and the other is polar molecules. For a non-polar dielectric, the stronger the external electric field, the greater the induced dipole moment, the more surface polarization charges, and the stronger the polarization of the dielectric. For dielectrics composed of polar molecules, the process of generating polarization is different from the above. Although each molecule has a certain inherent dipole moment, in the absence of an external electric field, the molecules do not exhibit electrical properties to the outside world because the molecules undergo chaotic thermal motion. However, under the action of an external electric field, each molecule is subject to an electric field torque. Under the action of this torque, the short dipole of the molecule will turn to the direction of the external electric field. For the entire dielectric, polarization charges are still generated on both surfaces perpendicular to the direction of the electric field. To sum up, although the microscopic mechanisms of polarization of different dielectrics are different, macroscopically, they all manifest as the appearance of surface polarization charges on the surface of the dielectric or the appearance of body polarization charges inside the dielectric, that is, the polarization phenomenon occurs. This polarization phenomenon is the root cause of insulation degradation and corrosion caused by no-clean flux residue after welding. In addition, high temperature and high humidity will also intensify the polarization phenomenon. Although the no-clean fluxes currently on the market have low solid content and are formulated to minimize the corrosiveness of their active ingredients, dielectric residues on the printed circuit board after soldering cannot be completely ruled out. Therefore, for circuit boards that work under hot flash conditions for a long time, insulation degradation and corrosion will occur between lines under the action of electric fields.

 

 

Reliability evaluation test

 

At present, the most commonly used reliability evaluation test in China is the surface insulation resistance test. The test method is as follows: During the test, use a comb-type electrode or a ring-type electrode of the specified material, evenly coat a certain amount of flux, and dry it for 30 minutes at a temperature of about 85°C to serve as a test piece. First, measure the insulation resistance of the above test piece under normal conditions, and then place the test piece in a constant temperature and humidity box with a temperature of 40°C ± 2°C and a humidity of about 90%. Keep it for 96 hours, take it out, and then put it into a test chamber for 20 hours. Put the saturated solution of special-grade sodium tartrate at a temperature of °C±°C into a dryer with an adjusted humidity of 90%, take it out within 1 hour, and then use an insulation resistance tester to measure the surface insulation resistance under standard conditions. Can this conventional test method be used to accurately evaluate the reliability of no-clean flux? According to the data, it can be seen that foreign countries have higher requirements for the surface insulation resistance of no-clean flux, and generally require bias voltage and long-term hot flash tests. Observe the aging effect of post-soldering flux residue on surface insulation resistance to measure the reliability of no-clean flux. Some problems can be illustrated through the following series of experiments. 2.1 Test method The test uses a comb-type electrode as shown in Figure 1. Clean the comb electrode with alcohol and dry it fully, apply a certain amount of no-clean flux on it, and put it into a furnace at a temperature of 235°C for 5 seconds after drying to prepare a sample. Then put the sample into a constant temperature and humidity chamber with a temperature of 40°C±2°C, a humidity of 95% RH, and an external bias voltage of 100V. At 96h, 150h, 200h, and 500h, take out the sample from the constant temperature and humidity chamber for 1h (the time required to fully dry it), and measure the insulation resistance value of the sample when the measurement voltage is 5V. After the measurement, quickly put it back into the constant temperature and humidity chamber to continue the test. Number of samples: n=6. 2.2 Test results Using the above test methods, three types of no-clean fluxes were tested. Table 1 lists the average surface insulation resistance corresponding to three no-clean fluxes. It can be seen from Table 1 that the surface insulation resistance values of the three no-clean fluxes have no significant changes between 96 and 200h. However, a downward trend appeared after 300 h, and the downward trend accelerated significantly as time extended.

 

 

Table 2 illustrates that the surface condition of the no-clean flux also changed after the voltage hot flash test. Moreover, as time increases, the surface conditions of the samples change to varying degrees. Among them, sample #3 showed obvious corrosion phenomena. The surface of the sample was covered with green spots and the copper wire of the sample was rusted and thinned. Figure 2, Figure 3, and Figure 4 show the surface condition of the sample after 500 hours of hot flash test.

 

 

 

 

surface condition  Time t/h                                     1#                                          2#                                          3#

 

96

                                        No discoloration of copper wire      No discoloration of copper wire   No discoloration of copper wire

 

150                                  Copper wire color deepens              Green rust spots appear            The copper wire is slightly blackened

 

200                                  No change                                      Green spots become larger          No change

 

From Figure 2 and Figure 3, it can be seen that the surface conditions of the l# and 2# samples did not change significantly after 500 hours of constant temperature and humidity, and bias voltage testing. However, it can be seen from Figure 4 that sample #3 has undergone major changes, leaving obvious rust spots and stains on the surface of the sample, and the copper wire is significantly thinner than that of other samples. This shows that the post-welding residue of this no-clean flux is highly corrosive under long-term load conditions in the humid zone.

 

3 Conclusion

 

(1) The reliability evaluation of no-clean flux is very important. It will be related to the overall reliability of electronic products. (2) The reliability test of no-clean flux must consider its aging effect. The actual application environment of electronic products using no-clean flux must be considered. ​

 

(3) Post-soldering constant temperature and humidity plus bias voltage reliability evaluation test is essential for no-clean flux.