Paul Feeney from Axus delivered an invited talk at the Clarkson University 19th annual symposium on CMP. The talk first gave an overview of the tremendous amount of activity that is taking place today to push traditional scaling for logic and memory devices. All the changes to materials and structures to make leading edge transistors had led to much more planarization being utilized for transistor formation. A focus for CMP in the FEOL is on combining replacement metal gates and multi-gate channels. For the interconnects, there is increased emphasis on controlling resistance effects as they come to dominate overall device performance. The interconnect section of the International Technology Roadmap for Semiconductors (www.ITRS.net) covers a number of things that need improvement to support future semiconductors. The net of more CMP applications, more average steps per application, and more interconnect layers is a steady rise in the number of CMP steps needed to produce a chip.
Meanwhile, the ITRS is taking a new direction. Increasingly, all the new types of devices that fall outside of traditional scaling, referred to as More than Moore, are being looked to in order to drive growth for the industry. This includes sensors, energy harvesting, lighting, etc. Going forward, the ITRS will turn its focus to these areas.
As all this is happening, the way people around the world come together to satisfy all the planarization needs has also been changing. The large manufacturers will add more CMP and thinning steps, but without adding significantly more resources. They are increasing their reliance on the equipment and consumables suppliers for more complete and mature CMP process solutions. As the equipment suppliers apply greater resources in support of their larger customers, there are likely to be fewer resources available to support the smaller device makers. This creates a challenge for the smaller manufacturers that are doing the majority of the work in the More than Moore space. Where will these operations turn to for their support?
As a part of this waterfall effect of needs, a growing number of CMP and grinding support operations are picking up the slack. Universities and development/foundry companies are increasing the breadth and sophistication of their offerings in order to provide process engineering and production expertise to the growing number of mid-size and entrepreneurial operations that need expert support. As it turns out, this is where Axus Technology puts its major focus. And not by accident, another key focus area for Axus is in supporting the consumables suppliers who often cannot afford their own development lab with the sophistication to help them deliver the needed complete and mature solutions.
Click here to view Paul’s presentation from the CAMP 2014 meeting.