Issue link: https://iconnect007.uberflip.com/i/1140547
JULY 2019 I DESIGN007 MAGAZINE 81 quality of the drilled microvias can be readily examined. Since this was just the first phase feasibility study, no statistical study was per- formed on the dimensional and positional tol- erances of the drilled microvias. The next step is to print the conductive ink traces on the top and bottom layers of the sub- strates. It was expected that, by printing the ink traces, the microvias would be automat- ically filled by the conductive ink. However, the results did not turn out as expected. Figure 10 shows the unfilled microvias after ink print- ing. When a liner was not used beneath the substrate, the vacuum sucked the conductive ink out of the microvias and the conductive ink formed a splattering pattern around the via holes on the bottom layer of the substrate (Fig- ure 10a). Even when a liner was used under the substrate, microvias were not filled either (Figure 10b). This phenomenon occurred for both 250 µm and 150 µm microvias. A number of trials were performed in order to fill the microvias and understand the root causes for the unfilled microvias. By analyzing the characteristics of the conductive ink used in this study and trial results, we gradually formed a list of potential root causes. The vis- cosity of the conductive ink used to print the traces, 20,000 cps, is too low for microvia fill- ing. For via filling in the PCB industry, a typi- cal value for the viscosity of the conductive ink is in the range 40,000 cps. The low viscos- ity of the ink makes it difficult to stay inside the via holes. Since we use a screen to print the conductive ink, the fine wires of the screen may drag the ink out of the holes during the release of the screen after ink printing. Some ink may stay on the liner if there is a certain adhesion between the ink and the liner when the liner is separated from the substrate. Oth- er causes may include via hole size effect and substrate material/thickness effect. Based on this analysis, we defined our plan to address the unfilled microvia issue. At first, we planned to evaluate conductive adhesive with higher viscosity for hole filling using stencil printing instead of screen printing. Sec- ondly, we planned to evaluate a new liner with weak adhesion to the conductive ink. Silicone film has been shown to have weak bonding with typical conductive ink. We also planned to work with our partners to evaluate different holes sizes from 500 µm down to 100 µm on another substrate material, such as TPU. After several trials, we successfully plugged the microvias using stencil printing conduc- tive adhesives. Figure 11 shows the filled mi- crovias 150 µm in diameter on the top and bot- tom sides. After filling the microvias, ink traces were printed on the top and bottom sides of the sub- strates. Figure 12 shows the printed substrate sample with printed circuitry on the top and bottom sides. Figure 9: Measurements of the drilled microvias (250 µm). Figure 10: Ink splattering on the bottom of substrate (a) and unfilled microvias (a/b). Figure 11: Filled microvias: (a) top side, (b) bottom side.