In with the … Old? Print E-mail
User Rating: / 0
Written by E. Jan Vardaman   
Friday, 28 September 2012 15:02

At this year’s IMAPS came the question, is today’s technology “new,” or just a makeover?

At the IMAPS International Symposium on Microelectronics in September, several speakers referred to technologies that are reemerging from developments introduced years ago. Presentations focused on topics such as flip-chip and 3D packaging with through silicon vias (TSVs), and the emerging favorite, die mounted on silicon interposers with TSVs.

In a thought-provoking keynote on three-dimensional integration, Dr. Subu Iyer, an IBM fellow, reminded the audience that we can trace the roots of what is being called 2.5D today back to the multichip modules introduced in the 1970s, when IBM used ceramic substrates with multiple flip-chip die. The technology migrated to thin-film on glass ceramic substrates also populated with flip-chip die. Today’s silicon interposer recalls the silicon interposer technology introduced during the 1990s. Thin-film-on-silicon MCM-D, where “D” stood for “deposited,” was one substrate choice, competing with laminate (MCM-L) and ceramic (MCM-C) flavors. At least 16 companies developed thin-film-on-silicon technology or products. Intel also developed a silicon interposer technology and established a pilot line.

The drivers cited today are similar to those listed 20 years ago and include improved performance, partitioning large die, integrating single chips into a module, and reducing die size where substrate density is the constraint. Additional advantages today include use of a silicon interposer to reduce the chip package interaction (CPI) problem. Using a substrate that is TCE-matched to the die (silicon) could minimize stress on the chip interlayer dielectrics and provide improved reliability.

Iyer spoke of the value that the silicon interposer offers where decoupling capacitors can be  integrated into the thin film layers on the silicon substrate. This is not a new concept: Silicon interposers with integrated passives have been in production for several years in RF modules. NXP’s spin-off, IPDiA, manufactures silicon interposers for RF modules, and hundreds of millions of units have been produced. Intarsia, a startup company funded by Dow Chemical, developed many integrated passive solutions that resulted in a large portfolio of patents, most of which have expired. STATS ChiPAC offers modules today that incorporate integrated passives into the thin-film-on-silicon and the number of designs using the technology is increasing.

Dr. Rao Tummala, formerly from IBM, now director of Georgia Tech’s 3D Systems Packaging Research Center, offered glass as an alternative to settle the need for a lower-cost interposer solution. An IMAPS session contained a number of papers devoted to glass interposer developments, including the fabrication of vias in glass. Glass as a substrate is not a new idea, either. STMicroelectronics has shipped millions of internally produced thin-film-on-glass transceiver/receiver modules. What is new is the formation of TSVs in the glass interposer.

Whereas many of the technology developments are not completely new, issues associated with the technologies are also familiar. These challenges include availability of design tools, known good die solutions, test methodologies and strategies, and the need for a well-defined assembly infrastructure.

Flip chip returns to copper. The first flip chip produced at IBM in 1964 used a copper ball as the bump. While flip-chip technology migrated through different solder alloys with a variety of bump fabrication methods from evaporation to printing and finally plating, much of the industry is currently moving to a copper pillar technology. Today's copper pillar typically has a solder cap, but it is ironic that flip chip has migrated back to copper after 48 years. The difference today is that the bump pitch is much tighter and the chip carrier is not necessarily ceramic.

Lessons from the past. In many of today’s technology introductions, the industry may benefit from lessons from the past. As a former University of Texas basketball coach recently stated, when developing a successful basketball program, it is often good to find players that have experienced failures. The same may apply to a technology program. Taking advantage of the lessons from the MCM and silicon interposer era, including commercial failures, may provide a better chance for success this time.

Silicon interposers (MCM-D) technology did not penetrate the high-volume commercial market and lost the battle to organic laminate structures. Essentially, the silicon substrates were excluded because of cost and logistics (including a supply base). The competing laminate substrates were less dense due to the rough topography of the laminates, but had much lower cost. From a logistics standpoint, the higher density of the silicon substrate could not be fully utilized because the large number of high lead-count die that it could support created a package that was extremely difficult to test and had low yields. Technologists with battle scars from the MCM-era may find a new team that needs their skill and experience. Perhaps with new management, the technology will become more than a niche package.

E. Jan Vardaman is president of TechSearch International (; This e-mail address is being protected from spambots. You need JavaScript enabled to view it . Her column appears bimonthly.

Last Updated on Monday, 01 October 2012 13:12


Eastern-US: China’s New Competitor?

Parity emerges among EMS Factories from Asia, Mexico and the US.

For the first time in years we see parity in the Eastern US among EMS factories from Asia, Mexico and the US. This EMS market condition will permit American OEMs (the EMS industry refers to OEMs as customers) to have more EMS pathways to choose from. Now more than ever, such EMS assignments will require deeper investigation relating to the OEMs’ evaluation of manufacturing strategies.

The Human Touch

For those who count on the electronics industry for big feats, it’s been a remarkable couple of years.



Advances in Concentration Monitoring and Closed-Loop Control

Contaminated bath water skews refractive index results. New technology can accurately measure aqueous cleaning agent concentration.

Circuits Disassembly: Materials Characterization and Failure Analysis

A systematic approach to nonconventional methods of encapsulant removal.





CB Login



English French German Italian Portuguese Russian Spanish


Techspray Introduces Fine-L-Kote High Viscosity AR Conformal Coating
Fine-L-Kote high viscosity AR acrylic conformal coating reportedly widens the process window and flexibility. Can use as-is for dipping or thin down for spray systems. Is for selective spray systems...