Advanced Packaging - From the Editor

It's interesting that in the space of half an hour at the same symposium — IMAPS International 2008 — I could experience such vastly different answers to the same question. It just further solidified the notion that perspectives differ on emerging technologies, even within the same industry. The question was about through silicon vias (TSVs), and what was holding up market adoption. Don Ream, sales and marketing manager at ASM Pacific, told me quite frankly that TSVs "are a pipedream" and he has serious doubts about their adoption, ever. He says that they used to say the same thing about wire bond technologies when flip chip was first developed, and that wire bond would be replaced by flip chip. It's taken 40 years for flip chip to become mainstream, and wire bond still exists.

30 minutes later, on the other side of the room, Sergey Savastiouk of All Via wasn't surprised to have heard that opinion, but assured me that TSV is already in production for a variety of products. However, he agreed with Ream's comparison to the early days of flip chip development. He says adoption is following a similar path of flip chip, but suggests that it will happen in an accelerated timeframe. It took a major IDM like Intel to launch flip chip into mainstream, and Savastiouk predicts it will most likely take a similar situation for TSV to make the leap. In the mean time, valuable reliability data is being gathered to help the technology through the next step. He eleborates in his industry forum piece appearing in the December 2008 issue of Solid State Technology, our sister publication.

One difference I see between the timelines of flip chip adoption and TSV adoption is that no one ever suggests TSV will replace flip chip or wire bond processes all together, but rather, will join them where suitable as a viable alternative in specific applications. The "Holy Grail" is a monolithic super package that incorporates all available interconnect technologies at their most reliable, efficient, and cost-effective usefulness.

As the IMAPS show drew equipment and materials manufacturers largely focused on high-frequency, RF, biomedical, aerospace, automotive, and military markets, exhibitors were more than likely not be early adopters of TSVs. There are still an abundance of breakthrough technologies in the flip chip and wire bond markets, and advancements in flexible circuits are grabbing the limelight for medical electronics, not to mention the new kid on the block, photovoltaics.

In any case, it was apparent that while IMAPS technical papers look to the future, exhibitors focused on specific and current market needs, and their new offerings to address those needs. So, depending on where you sit, both responses to my original question were correct. The markets served by ASM Pacific, Reims' company, may never be adopters of TSV. And, with any luck, a consumer electronics manufacturing mogul will push for the adoption of TSV in its market space, making all the work done by a plethora of materials and equipment manufacturers, research institutes, universities, and consortiums worth the effort. It all depends on your perspective.

Add RSS Feed