3D IC is only solution for scaling “up,” says MonolithIC 3D exec

By Debra Vogler, senior technical editor

March 31, 2011 — Serial entrepreneur, Zvi Or-Bach, was interviewed by ElectroIQ about his latest startup, MonolithIC 3D. In this podcast interview, Or-Bach explains why the company changed its name (from NuPGA) when its mission changed. "The name change was in response to a strategic change we made once we discovered a path to monolithic 3D ICs," said Or-Bach. He noted that on-chip interconnects are the limiting aspect with respect to scaling, and that after making materials changes (i.e., Al to Cu, and SiO2 to low-k), the only solution is to go to 3D ICs.

Listen to Or-Bach’s interview:  Download (iPod/iPhone users) or Play Now

Interconnect delay → Monolithic 3D   Implant H+ dummy gates Transfer on top of processed wafer and replace gates (<400°C)
  90nm (2005) 45nm (2010) 22nm (2015) 12nm(2020)
Transistor delay 1.6ps 0.8ps 0.4ps 0.2ps
Delay of 1mm-long interconnect 5 x 102ps 2 x 103ps 1 x 104ps 6 x 104ps
Ratio 3 x 102 3 x 103 4 x 104 3 x 105
3D 

Figure. Next-generation monolithic 3D IC — leveraging the gate-last process. SOURCE: MonolithIC 3D

Semiconductor scaling costs

Scaling down 0.7x Scaling up (3D packaging)
 Cost: Capital >$4B  Cost: Capital: <$200M
   R&D >$1B   R&D  <$100M 
 Benefits: Die size 0.5x Benefits:  Die size 0.5x 
  Power 0.5x   Power 0.5x
Speed: No change Speed: No change

Table. The next-generation dilemma — going up or going down? Companies can do both. SOURCE: MonolithIC 3D. 

"While TSVs are a big help with off-chip interconnects, they are not helpful for on-chip interconnects — they are just too large," explained Or-Bach. "Our vertical interconnect is 10000× more dense than TSVs." The company’s mission is an answer to what Or-Bach calls the next-generation dilemma (table). Whether one chooses to continue to scale down 0.7×, or scale "up" by going to 3D, there are costs. However, the company’s estimates show a glaring difference: capital costs and R&D costs are, respectively, >$4B and >$1B for scaling down; and <$200M and <$100M for scaling up.

3D IC technology

The technology being proposed by Or-Bach uses a combination of four ideas:
1) the gate-last process and proper sequencing of ion-cut (i.e., Smart Cut) technology;
2) low-temperature face-up layer transfer;
3) repeating layouts;
4) innovative alignment.

Or-Bach explains the process in detail (figure) in his audio interview. The technology still requires process development work, but looking ahead, Or-Bach views it as being applicable to both Tier 2 fabs that want to reinvent themselves and compete with leading-edge fabs, and leading-edge fabs that want to add value.

Subscribe to Solid State Technology/Advanced Packaging.

Follow Advanced Packaging on Twitter.com by clicking www.twitter.com/advpackaging. Or join our Facebook group

POST A COMMENT

Easily post a comment below using your Linkedin, Twitter, Google or Facebook account. Comments won't automatically be posted to your social media accounts unless you select to share.

3D IC is only solution for scaling “up,” says MonolithIC 3D exec

By Debra Vogler, senior technical editor

March 31, 2011 — Serial entrepreneur, Zvi Or-Bach, was interviewed by ElectroIQ about his latest startup, MonolithIC 3D. In this podcast interview, Or-Bach explains why the company changed its name (from NuPGA) when its mission changed. "The name change was in response to a strategic change we made once we discovered a path to monolithic 3D ICs," said Or-Bach. He noted that on-chip interconnects are the limiting aspect with respect to scaling, and that after making materials changes (i.e., Al to Cu, and SiO2 to low-k), the only solution is to go to 3D ICs.

Listen to Or-Bach’s interview:  Download (iPod/iPhone users) or Play Now

Interconnect delay → Monolithic 3D   Implant H+ dummy gates Transfer on top of processed wafer and replace gates (<400°C)
  90nm (2005) 45nm (2010) 22nm (2015) 12nm(2020)
Transistor delay 1.6ps 0.8ps 0.4ps 0.2ps
Delay of 1mm-long interconnect 5 x 102ps 2 x 103ps 1 x 104ps 6 x 104ps
Ratio 3 x 102 3 x 103 4 x 104 3 x 105
3D 

Figure. Next-generation monolithic 3D IC — leveraging the gate-last process. SOURCE: MonolithIC 3D

Semiconductor scaling costs

Scaling down 0.7x Scaling up (3D packaging)
 Cost: Capital >$4B  Cost: Capital: <$200M
   R&D >$1B   R&D  <$100M 
 Benefits: Die size 0.5x Benefits:  Die size 0.5x 
  Power 0.5x   Power 0.5x
Speed: No change Speed: No change

Table. The next-generation dilemma — going up or going down? Companies can do both. SOURCE: MonolithIC 3D. 

"While TSVs are a big help with off-chip interconnects, they are not helpful for on-chip interconnects — they are just too large," explained Or-Bach. "Our vertical interconnect is 10000× more dense than TSVs." The company’s mission is an answer to what Or-Bach calls the next-generation dilemma (table). Whether one chooses to continue to scale down 0.7×, or scale "up" by going to 3D, there are costs. However, the company’s estimates show a glaring difference: capital costs and R&D costs are, respectively, >$4B and >$1B for scaling down; and <$200M and <$100M for scaling up.

3D IC technology

The technology being proposed by Or-Bach uses a combination of four ideas:
1) the gate-last process and proper sequencing of ion-cut (i.e., Smart Cut) technology;
2) low-temperature face-up layer transfer;
3) repeating layouts;
4) innovative alignment.

Or-Bach explains the process in detail (figure) in his audio interview. The technology still requires process development work, but looking ahead, Or-Bach views it as being applicable to both Tier 2 fabs that want to reinvent themselves and compete with leading-edge fabs, and leading-edge fabs that want to add value.

Subscribe to Solid State Technology/Advanced Packaging.

Follow Advanced Packaging on Twitter.com by clicking www.twitter.com/advpackaging. Or join our Facebook group

POST A COMMENT

Easily post a comment below using your Linkedin, Twitter, Google or Facebook account. Comments won't automatically be posted to your social media accounts unless you select to share.