Page 40 - Chip Scale Review_January February_2023-digital
P. 40

Figure 8: Crack probability plot for: a) (left) pad stack option A vs. b) (right) B and tip diameter 5µm on the Weibull scale.
        burst signal energy lower than 5eu are   burst signal energy and selecting only   and the y-axis to ln [ln[1/(1-P f )]. The
        considered as noise, friction, or plastic   those hits that have occurred for the first   Weibull parameters m and F 0  are extracted
        deformation of metal layers below the   time per indent (“first oxide crack”). The   from a linear regression line that is fitted
        oxide layer, which are noncritical and not   cumulative crack probability P f , which   through the data points [2].
        relevant for the crack assessment.  follows a Weibull distribution with the   Figure 8 shows the graphical result for
          The critical contact forces can be   Weibull modulus m and the characteristic   a sample size of 150 indents using a 5µm
        statistically derived by filtering the data   contact force F 0 ,, can be graphically   flat punch tip (FP05) each for pad option A
        from the scatter plots above a certain   linearized by scaling the x-axis to ln(F)   (left graph) and B (right graph) in Weibull
                                                                              scale format. Both graphs show a black
                                                                              fitting line that has been adapted to the
                                                                              data points. By definition, the data analysis
                                                                              corresponds to the characteristic contact
                                                                              force F 0  with a 63% crack probability
                                                                              and confirms the theory that the critical
                                                                              forces causing cracks in brittle materials
                                                                              follow a Weibull distribution. Through
                                                                              extrapolating the regression line of the
                                                                              crack probability model, the crack risk at
                                                                              low ppm-rates can be determined, which
                                                                              is later important for complying with the
                                                                              POAA quality requirements.
                                                                                Table 1 shows the results of the crack
                                                                              probability assessment for all combinations
                                                                              of pad options A and B using two different
                                                                              tip diameters. All experiments were done
                                                                              for a sample size of 150 indents.
                                                                                After completing all experiments and
                                                                              data analysis, an assessment to identify
                                                                              the best pad/tip diameter combination can
                                                                              be visualized (see Figure 9). Comparing
                                                                              the four graphs, it can be seen that a two-
                                                                              times larger tip diameter (10µm vs. 5µm)
                                                                              increases the critical force up to a factor of
                                                                              2.5. It is important to note that the slope of
                                                                              the regression lines, which corresponds to
                                                                              the Weibull parameter m, is steeper for pad
                                                                              option B. This can be interpreted as a lower
                   E-Tec Interconnect  AG, Mr. Pablo Rodriguez,  Lengnau Switzerland  crack risk for low crack probability rates. In
                       Phone : +41 32 654 15 50, E-mail: p.rodriguez@e-tec.com  conclusion, pad option B is approximately
                                                                              35% more robust compared to option A,


        38
        38   Chip Scale Review   January  •  February  •  2023   [ChipScaleReview.com]
   35   36   37   38   39   40   41   42   43   44   45