The transition for Cu barrier metal from Ta to Ru
seems to have been once again delayed (after 22nm node). Is anyone currently
working on Ru integration, and is there more work needed on the CMP step?
We continue to see interest in Ru barrier along with growing interest in Co. Last year we (Richardson ICPT Taiwan Nov. 2008) reported a dual approach of selective and non-selective Ru slurries depending on whether the integration schemes need a hard stop on oxide or not. Ru has both issues and opportunities-- i.e. challenges in deposition and cost, low contact resistance, good adhesion. And as usual, the conservatism of the industry dictates a preference to extend older technologies whenever possible. In my opinion, Ta will not always meet the needs at 22nm or lower pitch.
Thanks for the update Cliff. From what you've written, it seems that you feel that the slurry set(s) could be ready today if needed. Without disclosing any confidential information, can you give us an indication of which application (Cu dual-damscene for interconnect, or DRAM capacitor liner, or ???) seems the most likely to call for Ru at 22nm?
Regarding #1 Happy New Year all! I've hear of some people considering PVD rather than ALD or CVD Ru ... in your opinion, will that effect the slurry or CMP process?
Actually Ed, IBM reported at CAMP and ICPT the use of Ru liner for copper contacts that replace tungsten plugs. So the application is neither interconnect nor capacitor! Not that these aren't possible.
Replying to Karey (Post #3, above), I recall seeing some work published by Cabot a year or two ago regarding the experience that the Ru CMP process indeed required tuning to match the deposition method. However, I do not recall if anything was quantified...
Cliff, is there anything you can share regarding Ru CMP optimization for ALD vs. PVD vs. CVD (vs. MOCVD vs. MVD vs. etc.)?
One of the things I learned at the 2009 ECS & MRS meetings in San Francisco was that the metallurgy itself is not yet necessarily stable, i.e. lots of combinatorial work is underway on best barrier alloy compositions with Ru. Until that settles down to a finite number of de facto industry standards, CMP will once again find itself chasing a moving target. That would be tremendous job security if it weren't for the profits that are critical to fund the development work.
Mike, I remember chatting with you at MRS09 about CMP...and you're (of course) right that the question of optimum composition has not yet been established, on top of which we must consider subtleties of nanostructure (grain sizes and distributions, dopants/contaminants, interfaces, etc.) which depend upon deposition method.
As always, it's better (cheaper, faster, safer) to extend a POR instead of integrating something new...as long as you can!
Maybe IBM (and associated "ecosystem partners"...a.k.a. the entire world except for Intel and TSMC) will use Ru along with Cu plugs at 32nm. If so, then the reason to go to Ru is probably the ability to reduce Cu resistance. Note that NEC reported on this effect at IITC2007 (Ref: http://www.pennwellblogs.com/sst/eds_threads/2007/06/070608-iitc2007-airgaps-chip-stacks.php):
"NEC research labs showed that direct ECD of Cu without a Cu-seed
layer provides larger grain size and higher Cu(111) orientation.
Damascene structures were first sealed with TiN, then either Ta/Cu or
Ru layers were deposited. The TiN barrier layer is definitely needed
beneath Ru to block Cu diffusion into the dielectric. Ru PVD using DC
magnetron sputtering with Ar gas at room temperature produces high
orientation of Ru(002). Since Ru(002) is hexagonal-close-packed, it
matches well with the preferred Cu(111) face-centered-cubic orientation
such that 40%-50% can be grown directly on Ru in dual-damascene
structures. Some day, metal line specifications may include not just
dimensions and resistivity, but grain orientation and size-distribution
too."
We're now (January 20th, 2010) past the
official ending (the 18th) of this virtual roundtable discussion, after 4695 views of 129 replies to 18 questions. I'll
edit together Interesting discussions from most of the topics into a
summary document that will be posted to the Planarization Lounge.
We'll leave the topic posting open in case there are additional comments...but they would not be included in the summary.
Regarding #1 Happy New Year all! I've hear of some people considering PVD rather than ALD or CVD Ru ... in your opinion, will that effect the slurry or CMP process?
Replying to Karey (Post #3, above), I recall seeing some work published by Cabot a year or two ago regarding the experience that the Ru CMP process indeed required tuning to match the deposition method. However, I do not recall if anything was quantified...
Cliff, is there anything you can share regarding Ru CMP optimization for ALD vs. PVD vs. CVD (vs. MOCVD vs. MVD vs. etc.)?
Mike, I remember chatting with you at MRS09 about CMP...and you're (of course) right that the question of optimum composition has not yet been established, on top of which we must consider subtleties of nanostructure (grain sizes and distributions, dopants/contaminants, interfaces, etc.) which depend upon deposition method.
As always, it's better (cheaper, faster, safer) to extend a POR instead of integrating something new...as long as you can!
Maybe IBM (and associated "ecosystem partners"...a.k.a. the entire world except for Intel and TSMC) will use Ru along with Cu plugs at 32nm. If so, then the reason to go to Ru is probably the ability to reduce Cu resistance. Note that NEC reported on this effect at IITC2007 (Ref: http://www.pennwellblogs.com/sst/eds_threads/2007/06/070608-iitc2007-airgaps-chip-stacks.php):
"NEC research labs showed that direct ECD of Cu without a Cu-seed layer provides larger grain size and higher Cu(111) orientation. Damascene structures were first sealed with TiN, then either Ta/Cu or Ru layers were deposited. The TiN barrier layer is definitely needed beneath Ru to block Cu diffusion into the dielectric. Ru PVD using DC magnetron sputtering with Ar gas at room temperature produces high orientation of Ru(002). Since Ru(002) is hexagonal-close-packed, it matches well with the preferred Cu(111) face-centered-cubic orientation such that 40%-50% can be grown directly on Ru in dual-damascene structures. Some day, metal line specifications may include not just dimensions and resistivity, but grain orientation and size-distribution too."
We're now (January 20th, 2010) past the official ending (the 18th) of this virtual roundtable discussion, after 4695 views of 129 replies to 18 questions. I'll edit together Interesting discussions from most of the topics into a summary document that will be posted to the Planarization Lounge.
We'll leave the topic posting open in case there are additional comments...but they would not be included in the summary.
Happy planarizing.