2015 overall was a mixed year for the materials sector in the electronics market, with a general slowdown in PC-related segments offset by continued growth in mobile smart phones and other handheld devices despite industry growth flattening out late in 2015. In 2016, market forecasts indicate some improved growth and potential technology shifts.
Looking at the 3D IC packaging market specifically, there seems to have been a significant shift over the past couple of years. Not too long ago, every conference was dominated by discussions of TSVs as the next 3D technology. And certainly TSVs are being implemented in some areas, however, cost remains a concern. Now, conferences are dominated by discussions of fan-in to fan-out wafer-level packaging (FOWLP) technologies. In fact, at the recent SEMICON Europa 2015, consultant predictions for FOWLPs showed a hockey stick upturn in demand from 2016 into the future.
Market signals support this assertion that there will be a significant migration from fan-in to FOWLP technologies as a method of conserving a high level of connectivity in a small package space. In the past, FOWLP was used only in niche applications. But now, some FOWLP may go mainstream into a significant portion of electronics devices starting in 2016. And, if they go mainstream in mobile and smart devices as predicted, that will be significant for materials suppliers and manufacturers with material selections and design approaches being made.
This coincides with challenges in continued shrinks and rising costs for front-end manufacturing. As front-end manufacturing continues to move into 14/16nm densities, cost advantages previously seen on shrink are now being challenged. When you consider Moore’s Law, the movement to fan-out provides a transition point where packaging technology provides increased performance improvements.
Semiconductor chip manufacturers are obviously aware of these changes, and these front-end players have become more involved in packaging. This is clearly changing the packaging landscape and shifting some of the decision points. For foundries, their involvement in the front-end process is a differentiator that they bring to packaging in comparison to the outsourced semiconductor assembly and test providers (OSATs).
So, how does this relate to FOWLP adoption? OSATs have had strategies in the FO space for quite some time and fan-out impacts 3D assembly. Some FO technologies may help push off adoption of 3D packages. However, front-end players have an advantage because they produce the chips in these packages. If packages present more challenges, then perhaps the whole 3D argument of who will take the risk is answered. Is it a chip or an interconnection problem?
If or when the whole fan-out wave takes off in 2016, what does that mean for 3D? Does it become important in just some specific areas, such as memory and server processor packages, for the next couple of years? Can 2.5D FO technology really carry us into the future? The outlook from 2016 forward could remain significant in terms of predicting the future for FOWLP and the extent of adoption.
The Internet of Things (IoT) market landscape has been a much different story than FOWLP. IoT market growth is definitely ramping, however, this growth isn’t necessarily going hand-in-hand with new technology development and a need for new materials. In many cases, manufacturers have been leveraging older technology at the 45nm node and above, which requires less technology in the packaging space. As IoT applications and technologies are optimized, higher-performing end devices will be needed, which in turn will drive the need for more technology in advanced packaging in that space.
Geo-political shifts should certainly be taken into account when predicting the outlook for electronics production in 2016. There was significant movement in China in 2015 following a government push to localize investment in the semiconductor industry. Additionally, the IoT will be important in China in 2016 and investment in packaging seems to be accelerating there, which we expect to continue in 2016 and beyond. Chinese localization may change the balance of global production of electronics in the coming years.
All things considered, from FOWLP to IoT to the increased pervasiveness of smart devices driving us towards an increasingly connected world, expectations for 2016 for advanced packaging and semiconductor materials remain high for Dow. The potential for technology advancements and market growth is certainly promising.