2.5D and 3D technologies have been shown to lead to improvements in device and system performance by virtue of reductions in latency and power, increases in band width, and reductions in form factor. The basic technologies needed for such integration have also been demonstrated al bet it not in high volume low cost manufacture. However, there are a number of questions/concerns in the industry when it comes using the established Si-interposer technology for 2.5D and traditional Cu-filled TSV technology for 3D stacking of devices in advanced silicon nodes. These issues/concerns include but are not limited to the overall cost structure of these technologies; the supplier base business models; the overall supply chain continuity, integration and flow; and product reliability that appear to be slowing down or at least affecting the adoption of this technology into the main stream. These issues are beginning to lead the industry to consider alternatives that could provide similar benefits in performance and integration to the established technologies of Si-interposers and TSV 3D. These alternatives include but are not limited to glass interposers and advanced organic substrates but without the complications of the established technologies discussed above. In this talk we will discuss the alternative technologies and compare them to the established technologies ultimately providing an overall landscape of the pro’s and cons for each with respect to the networking and system needs. Further status of each technology with respect to high volume manufacturing and what if anything is needed to advanced them toward high volume manufacturing will be discussed




	Dr. John Osenbach Fellow, Materials and Interconnect Technology LSI

Profile

Fellow , LSI company 2007-2012 Principle member of tech staff , Agere 2002-2007 IC package and integration technology Fellow , Lucent Bell Labs 1999-2002 optoelectronics interconnect and integration Consulting member of tech staff AT&T/Lucent Bell Labs 1996-1999 optoelectronics interconnect and integration Distinguished member of tech staff AT&T/Lucent Bell Labs 1996-1999 optoelectronics interconnect and integration Distinguished member of tech staff AT&T Bell Labs 1987-1991 IC technology and integration technology Member of tech staff AT&T Bell Labs 1982-1987 IC technology and integration technology 55 US patents issued and 20+ pending <120 papers and talks, 25+ invited talks Taught OPC technical course on optoelectronics reliability 1999, 2000 Associate editor of J Electrochemical Society 1988-1991 Co-chair dielectric division of electrochemical society symposiums 1986-1988 BS/MS/PhD in Ceramic Science and Engineering, Penn State Univ. 1979, 1980, 1982


Drivers and Constraints for Using 2.5D/3D Technology for Networking and System Devices

Abstract

2.5D and 3D technologies have been shown to lead to improvements in device and system performance by virtue of reductions in latency and power, increases in band width, and reductions in form factor. The basic technologies needed for such integration have also been demonstrated al bet it not in high volume low cost manufacture. However, there are a number of questions/concerns in the industry when it comes using the established Si-interposer technology for 2.5D and traditional Cu-filled TSV technology for 3D stacking of devices in advanced silicon nodes. These issues/concerns include but are not limited to the overall cost structure of these technologies; the supplier base business models; the overall supply chain continuity, integration and flow; and product reliability that appear to be slowing down or at least affecting the adoption of this technology into the main stream. These issues are beginning to lead the industry to consider alternatives that could provide similar benefits in performance and integration to the established technologies of Si-interposers and TSV 3D. These alternatives include but are not limited to glass interposers and advanced organic substrates but without the complications of the established technologies discussed above. In this talk we will discuss the alternative technologies and compare them to the established technologies ultimately providing an overall landscape of the pro’s and cons for each with respect to the networking and system needs. Further status of each technology with respect to high volume manufacturing and what if anything is needed to advanced them toward high volume manufacturing will be discussed