Bubble generation during CMP

I'd appreciate any inputs and comments regarding bubble generation (and hopefully supression of bubbles) during CMP.

As you know metal polish slurries (especially copper and TSV) contain plenty of surfactants and hydrogen peroxide (other slurries do as well).

My experience has been that excessive bubble generation can be a real pain not only re: defects but also re: RR and overall pad surface quality management.

I am curious to know the community's thoughts on this.

Allow download:

Allow download

Ara Philipossian

01/12/2010 - 11:18am

# of views: 3926

# 1

Ed Korczynski

January 15,2010 - 3:34pm

Hmmm. As I type this, I see that 38 people have read this question over the last three days, with no replies yet.

I have to guess that in commercial fabs this problem is either uncommon (perhaps triggered by a subtle shift in parameters when trying to run near the limits of some non-standard process space), unknown (perhaps due to honest ignorance), or embarrasingly uncontrolled (but happily only causing a few % WIW and WTW nonuniformity) so no one wants to talk about it.

# 2

Ara Philipossian

January 17,2010 - 11:55am

Well - The number of views is at 49 now, but no replies other than Ed's.

I am surprised since foaming, frothing and bubbling all occur during CMP at varying extents depending on the type of slurry, tool configurations and tool settings. Maybe I have touched a sensitive chord and no one wants to talk about it.

I don't think that these phenomena are just nuisances or cosmetic issues and the CMP community needs to get a handle on this. Theoretically, these phenomena should affect the uniformity of the formation of the passivation films; they may also affect pad debris transport (through flotation depending on the contact angles involved), pad fouling and defect formation.

I do not have any data or else would have been happy to share them with you.

# 3

Rakesh Singh

January 18,2010 - 10:23am

It would have been nice to get some insight on this interesting topic from the fab/end users. The bubble generation during CMP process itself may not be a significant issue, and even not happening in reality, once a consistent flow rate of slurry is dispensed on the pad with numerous micro/macro cavities to be filled, and strong interaction with the pad conditioner, retaining ring and the wafer. In most applications, there may not be significant still time and consistent fluid volume/geometry to grow the microbubbles present in the slurry blend, and any slurry bubbles forming & growing must be minimal and not affecting the polishing uniformity beyond that caused due to the other factors. Further, this phenomenon being transient may not be very repeatable. On the other hand, avoidance of bubble formation during the slurry blending (especially the accurate metering of slurry component in some surfactant and other additive containing slurries) in the mixing vessels and daytank, distribution in the global loop (no dead legs), and finally accurate and consistent dispensing at the platen (using some type of pump) is very important and can be very challenging at times. Most of the latter issues may be resolved when proper care is taken during the above steps the slurry is periodically recirculated to get rid of any bubble accumulation in the parts of slurry system as well as the dispense lines.

# 4

Ed Korczynski

January 20,2010 - 5:48pm

Thanks, Rakesh, for addressing some of the fundamentals here.

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.