News & Analysis
Comment
Oscar Law
docdivakar
I agree... secondly one must realize that 3D IC design is also a packaging ...
Perfecting the 3-D chip
R Colin Johnson
10/11/2011 10:31 AM EDT
IBM + 3M = 3-D ICs. It’s a catchy formula. “Somewhere in the middle of those initials, 3M has the technology platforms to make 3-D work,” said Bernard Meyerson, vice president of research at IBM (Armonk, N.Y.)
After years of research into each of the component technologies needed to enable 3-D ICs, IBM decided there was a missing material so important that it cut a deal with 3M to create it. The key remaining obstacle to 3-D ICs, according to IBM, is an underfill material that can do double duty as an electrical insulator and a thermal conductor, wicking away heat from hot spots. IBM aims to use the material in conjunction with microfluidic channels containing refrigerants on 3-D structures.
“3M has the skills to meet the really disparate requirements for a 3-D IC adhesive,” said Meyerson. “You want infinite thermal conductivity in the adhesive, but you also want the electrical conductivity to be zero.”
The worst constraint, according to Meyerson, is that the coefficient of thermal expansion for the adhesive must match that of the metal used for the interconnect; otherwise, the adhesive will break the metallization when it heats up.
“Thermal conductivity, electrical conductivity and thermal expansion are all related, not to mention brittleness. It’s what we call an overconstrained system.”
Ming Cheng, technical director of 3M’s Electronics Markets Materials Division, said 3M “is essentially a materials company with the capability to tune the properties of adhesives and polymers to meet even these conflicting specifications. Our adhesive will be a combination of different types of polymers, oligomers and monomers, along with the necessary feelers and adhesion promoters that meet IBM’s specifications.”
According to 3M, it has not yet been decided whether the jointly developed 3-D IC adhesive will be sold to other chip makers. But in the past IBM has made a practice of licensing its key patents even to competitors.
3M also has experience with the fluids that are used today to cool hot spots in rack-mounted computers, and those fluids could soon flow through microfluidic channels cut into 3-D ICs. “Even if you have the perfect adhesive, it may be necessary to remove heat from the internal layers of a tall stack,” said Meyerson. “A microchannel-cooled radiator halfway through the stack could take out a bunch of heat from the middle of a silicon brick.”
Said Cheng: “Our Fluorinert electronic liquids are currently used to help cool equipment in data centers—mostly servers and hard drives—but with IBM we will also be exploring the liquids’ use to help cool 3-D ICs.”
Beyond perfecting the processing technologies to interconnect stacked dice and keep them cool, designers must consider the tide of data that will be streaming out of 3-D ICs, according to IBM. Photonics will become an integral part of 3-D ICs to handle the voluminous I/O.
“Electronic data transmission today can consume up to 50 percent of a chip’s power. Photonics is vastly more efficient in watts per bit and, for that reason, will be essential for 3-D ICs,” said Meyerson. “We’ll need lasers, modulators and detectors right in the 3-D IC stack.”
The work with IBM was announced just recently, but 3M has been working on 3-D solutions for some time. Earlier this year, in fact, 3M announced a technology for handling wafers destined for 3-D stacks. The company’s wafer-supporting-system (WSS) simplifies the handling of wafers that have been thinned for stacking.
WSS “first bonds the thin wafer to glass with a temporary adhesive so that the glass can support the wafer during bonding,” said Cheng. “After the two wafers are stacked, debonding allows the glass [carrier] to be removed.”
By 2013, 3M and IBM promise an end-to-end process ready for widespread high-volume commercialization of heterogeneous stacks of processor, memory, mixed-signal, networking and I/O chips formed into silicon skyscrapers as high as 100 chips per stack. — R. Colin Johnson



wbayer
10/11/2011 2:23 PM EDT
FYI, the Silicon Integration Initiative (Si2) also has a 3D standards effort under way, and is cooperating with the other entities listed in this excellent article to be sure there is no overlap.
For more info http://www.si2.org/?page=1380
Sign in to Reply
resistion
10/11/2011 6:48 PM EDT
The serial nature of stacking and the vertical thermal proximity are fundamental concerns.
Sign in to Reply
hm
10/11/2011 7:04 PM EDT
Long term reliability of these parts is also of concern. There has to be new testing methodology.
Sign in to Reply
Astronut
10/12/2011 12:19 PM EDT
hm: You're right! Although 3D itself shouldn't affect lifetime (none of our 3D-ICs from 2004 have failed yet), when more circuitry is added the mean time between failures is likely to become unacceptably small. Continual on-chip testing may be the answer.
Sign in to Reply
resistion
10/12/2011 2:30 PM EDT
Hope that can remain an option not become a specific requirement. :)
Sign in to Reply
chipmonk
10/12/2011 1:11 PM EDT
An excellent overview of current 3D status and roadmap except for the glaring omission of Allvia - a small foundry located right here in Si Valley that has pioneered TSV technology at a prototype level for the last 5-6 years and provides smaller customers access to this technology. They also have Si Interposers with integrated capacitors.
Sign in to Reply
JoeData
10/12/2011 1:49 PM EDT
Sounds like a perfect use for industrial diamonds - underfill material!
Sign in to Reply
chipmonk
10/13/2011 3:17 PM EDT
vapor deposition is being tried but cost remains an issue
Sign in to Reply
elctrnx_lyf
10/12/2011 2:19 PM EDT
will this technology be complete future in the next few years to much higher density processors on a single chip.
Sign in to Reply
resistion
10/12/2011 2:32 PM EDT
One also mustn't lose sight of 3d interconnect taking up more and more area especially with more and more stacked dies.
Sign in to Reply
chipmonk
10/13/2011 3:28 PM EDT
yes, the TSVs will take up about 15 - 20 % space on a die ( including the bond pads & keep-outs to isolate transistors from stress induced by the TSVs ) but this would be a wash since no large bond pads would be required for wire bonds any more. Because of the short I/O length ( low loss ) and high I/O density possible, TSVs will enable a lot of high - perf / low - power / compact designs.
Sign in to Reply
Astronut
10/13/2011 3:42 PM EDT
resiston: Many of the 3D interconnects won't go through the whole stack, just 1 or 2 layers. Only the signals to the outside world need to emerge at the surface.
Sign in to Reply
Rchandta1
10/12/2011 8:04 PM EDT
.."volume commercialization of heterogeneous stacks of processor, memory, mixed-signal, networking and I/O chips formed into silicon skyscrapers as high as 100 chips per stack .." - this is very ambitious goal!!
Generally sounds reasonable, but very skeptical of the idea of liquid cooling. May fine for special application but definitely not for commercial.
Sign in to Reply
ATUL SRIVASTAVA
10/13/2011 2:32 AM EDT
This will obviously need new generation of EDA tools too . What is the status on that front ?
Sign in to Reply
pinhead
10/13/2011 1:19 PM EDT
Good question. I was working for one of the cutting edge 3D companies a year ago, and at that point, the EDA was more or less non existent. It was just hacking things together manually.
Sign in to Reply
Astronut
10/13/2011 3:47 PM EDT
ATUL: Micro Magic, Magma, and R3Logic all have 3D-aware EDA tools; but there are still some gaps in the 3D toolset.
Sign in to Reply
docdivakar
10/13/2011 7:30 PM EDT
I agree... secondly one must realize that 3D IC design is also a packaging exercise so that needs to be addressed concurrently. Otherwise, there will be costly repititions and try outs of what is manufacturable.
MP Divakar
Sign in to Reply
collin
10/13/2011 11:51 AM EDT
I think TI is also developing 3-D IC now. If we have 3-D IC, does it mean that we could see 3-D video on TV, website,mobile phone,tablet,and all other devices which have display terminal?
Sign in to Reply
Astronut
10/13/2011 3:48 PM EDT
collin: 3-D video is a very different thing from 3D-ICs. 3D-IC refers to a way of designing and manufacturing chips; 3-D video is an application. No relation.
Sign in to Reply
docdivakar
10/13/2011 7:27 PM EDT
@Colin Johnson: other entities involved in 3D IC standardization is GSA which has been meeting 4 to 6 times a year on this topic, and Si2 Consortium.
IBM's work in 3D IC by stacking is nothing new, definitely not a secret. At CANDE 2010 and also at GSA 3D Standards meeting earlier this year, there were presentations by IBM folks so I am not at all surprised. What surprises me is 3M taking on the challenge to develop a thermal interface material with high conductivity for gaps less than 1um.
My earlier posting on 3D-related article is here:
http://www.eetimes.com/electronics-news/4215464/Si2-to-form-3-D-IC-standards-group?cid=NL_EETimesDaily
MP Divakar
Sign in to Reply
Oscar Law
10/17/2011 9:08 PM EDT
I think that there are still few challenges for 3D IC design:
- How can we test all IC before they are stacked together?
- Most of current 3D IC is limited to memory or sensor only, it takes time to develop complete strategy for stacking real ASIC together
- Wafer handling and KGD becomes nightmare for product engineer
- How can we handle thermal issues?
- 3M approaches may be good solution and we shall see how to apply this for production. However, I don't prefer liquid cooling solution, it is too complicated for implementation
- New 3D chip architecture related with TSV placement will solve the problem with lower cost
- Is new 3D IC EDA tools required or not?
- I prefer hybrid 3D IC design flow, it re-uses most of current ASIC flow with minor modification, it not only reduces the cost but also shortens the design cycle. For example, the multiple die physical verification time can be reduced by 10X or more through minor changes in current flow.
Sign in to Reply