Analysis of the interaction of various etchback factors of plasma in the PCB process 0 Preface The factors that affect the effect of plasma etchback include power, total gas, gas mixing ratio, etchback time, temperature and other factors. These factors not only directly affect the test results, but the interaction between the factors also affects the results. The so-called Interaction refers to the effect of the joint collocation of certain factors at various levels on the test results. This joint effect is called interaction. This article will explain and analyze the interaction of various factors of plasma based on the orthogonal test of etchback. The analysis results can provide reference for plasma etchback. 1 Etching test 1.1 Experimental equipment Plasma machine (Plasma Etch—Carson City.NV—Made in USA), 25 μm thick PI single-sided copper clad laminate, electronic analytical balance. 1.2 Experiment content In order to obtain the best etch-back effect, the factors of plasma etch-back test are selected, including power, total gas, gas ratio, and etch-back time. However, temperature is not selected for this test. The reason is based on the information and the author’s previous test experience. The etchback treatment does not require high temperature, so it is only necessary to control the test temperature within the normal range. It should be noted that in order to facilitate the investigation of the interaction between the factors, two test levels are selected for each test factor, which is simple and clear, and each test factor level is after a comprehensive analysis of the previous experimental results and data information. The ones that were chosen only have the characteristics of being neat, comparable and evenly dispersed. The selection of factor levels is shown in Table 1. As for the ratio of the two gases (carbon tetrafluoride and oxygen), it is the parameter of the best experimental effect of the author's previous cavitation, and only A1 and A2 are used here (Table 1).

According to the analysis of the previous test results, the factors that may interact in the plasma etchback test include gas ratio, total gas and power. Therefore, the interactions that need to be investigated are: gas ratio and total gas, gas ratio and power, and the interaction between total gas and power, namely A×B, A×C, B×C. Since the orthogonal test takes into account the interaction, when designing the orthogonal test table, the factors and the interaction that need to be investigated need to be reasonably arranged on the head of the orthogonal table, so the choice of the orthogonal table is particularly important. Generally speaking, the selection of the orthogonal table considering the interaction should follow a principle, that is, the sum of the degrees of freedom of the consideration factors and the degree of freedom of the interaction must not be greater than the total degrees of freedom of the selected orthogonal table. The calculation standard of the degrees of freedom is: the total degrees of freedom of the orthogonal table f total = the number of trials-1; the degree of freedom of the factors is f = the number of factor levels-1, and the degree of freedom of the interaction is the product of the degrees of freedom of the two interaction factors. In this experiment, since only two levels are arranged for each factor, according to the calculation criteria of the degrees of freedom above, the degrees of freedom of A, B, C, and D are: fA=fB=fC=fD=2-1=1 ; The degree of freedom of the interaction A×B, A×C, B×C is, fA×B=fA×fB=1, the same is true, fA×C=fA×fC=1, fB×C=fB×fC= 1. It can be seen that the sum of the degrees of freedom of the factors and interactions in this experiment = fA + fB + fC + fD + fA × B + fA × C + fB × C = 7, so you can choose the orthogonal table L8 (27) to arrange the experiment, because of the total freedom The degree is: ftotal=8-1=7, which just meets the principle of choosing the orthogonal table, so you can choose L8 (27) to arrange the experiment. At the same time, the design of the orthogonal table header should also follow the following two points: (1) First consider the interaction factors A and B, put A in the first column and B in the second column, and then query the L8 (27) interaction table to get A×B as the third column. (2) Consider the interactive C, and put C in the 4th column. From the L8 (27) interaction table, A×C occupies the fifth column, B×C occupies the sixth column, and D is the last column. Based on the above principles, an orthogonal table (Table 2) can be designed to arrange test Test. According to Table 2 to arrange 8 groups of tests, the tested test results, namely the amount of etchback, are represented by poor quality. The specific method is to compare the quality difference of single-sided copper clad laminates before and after the test (measured by electronic balance analysis). The quality difference represents Based on the etchback effect of plasma on PI, 8 sets of data were obtained by doing 8 experiments based on this standard. In order to scientifically analyze the influence of various factors and interactions on the experimental results, a range analysis of the various factors and interactions is also required. , And use this to judge which factors have a large impact on the experimental results, and those factors have a small impact on the experimental results. At the same time, you can also judge the factors that have a large impact on the experimental results. Which level will be more beneficial to the experimental results. 1.3 Analysis of experimental results The poor quality of the test results (the amount of etchback) and the range analysis are shown in Table 3. Now take A factor as an example to explain the principle and calculation of range analysis, divide the test results corresponding to the two levels of A factor into two groups, namely A1 and A2, and take the average of the two groups of test results: K1 = (0.694 + 1.508 + 0.953 + 0.963)/4 = 1.0295 K2=(1.295＋1.021＋0.893＋1.366)/4=1.14375 Then subtract the minimum value from the maximum value of K1 and K2 to get

Range value, namely: R=1.14375－1.0295=0.11425 The rest of the data can be deduced by analogy and then use the range analysis to get the result. From the results of the range analysis, it can be seen that a large range indicates that the two levels have a large impact on the experimental results, that is, it is usually an important factor, and the factor that causes the range is often an unimportant factor. We can conclude that the primary and secondary factors of this experiment are: (D: Time)-(C: Power)-(A×C)- (A: Gas ratio)-(A: Gas ratio)-(B: Gas Total)-(A×B)-(B×C) It can be seen that the influence of the interaction of factors in this experiment is still very large. The main influence is A×C, followed by A×B. As for the influence of B×C is too small to be ignored, then in this case , The selection of test parameters cannot be considered separately. For example, when there is no interaction in time, it can be considered separately as 35 min. The selection of the level of factors with interaction should be carried out by drawing a binary table and a binary graph. Comprehensive analysis, the binary table of interaction A×C is shown in Table 4, and the binary diagram is shown in As shown in Figure 1.

It can be clearly seen from the binary table and the binary graph that the level selection of interacting factors is the result of observing their interaction. The combination of A1C2 (C2 is 2600 W) makes the cavitation effect the most obvious. So choose A1C2. In the same way, we can determine the interaction A×B through the binary table (Table 5) and the binary graph (Figure 2): Since A has been determined by the previous experiment, that is, A1, the determination of B is relatively simple, in the A1 box In the selection, B1 should be selected, because A1B1 has the most obvious etchback effect.

3 Experimental conclusion From the analysis of the experimental results, it can be seen that in the plasma etchback test, there is an interaction between various factors. The main interaction that needs to be considered is the interaction between gas proportion and total gas, gas proportion and power. In practice, these interactions need to be fully considered and used to optimize production. Figure 25 shows a cross-sectional photograph of the bonding pad portion of the copper wire on the adhesive-free thick film circuit. Compared with the solder part, the thick-film conductor layer is very thin, and it can be understood that the thick-film conductor layer has a good affinity with the solder and can be in uniform contact.

4.5 Nano glue and copper glue For thick film circuits, the performance of printed adhesives for conductors is an important key point. In order to improve the conductivity or the clarity of the pattern, the particle size of the conductive particles is controlled within 100 μm, that is, the development of nano glue technology is popular. If you evaluate the performance of these nano-glues, the conductivity and the clarity of the graphics are indeed much higher than those of polymer thick film circuits. However, the cost of materials has also risen considerably, and a little carelessness will eat up the low cost of the printing process that has been difficult to obtain, and increase the ratio of materials in the cost. The result is counterproductive, and the cost is higher than that of the usual PCB formed by etching copper foil, and it is difficult to apply to consumer electronic equipment. In order to replace the expensive Ag conductor particles, inexpensive copper metal conductor particles have been tried. But this method is not perfect. This is because the copper fine powder is less stable than Ag and requires proper stability treatment, which will increase the cost and lose the advantages compared with Ag conductors.