Wafer broken

SangBum Kim kimsangb at stanford.edu
Fri Apr 10 16:59:13 PDT 2009


In my case, my oxide PECVD wafers (~1um thick) got broken while my TEOS wafers (~1.3um thick) did NOT. Both of them had not pattern on it. In addition, at that time, I was using contaminated metal set, so it could be a different factor as well.

You can tell whether the wafer got broken during pick up process or some other process by looking where your broken pieces are. If they are all over the polishing pad and, that means it got broken during polishing steps. If they are only found under the wafer head, it got broken after polishing steps when it is picked up.

Thanks,
SangBum

----- Original Message -----
From: "Chong Xie" <chongxie at stanford.edu>
To: "Christopher McGuinness" <cmcg at snf.stanford.edu>
Cc: "SangBum Kim" <kimsangb at stanford.edu>, cmp at snf.stanford.edu
Sent: Friday, April 10, 2009 2:31:17 PM GMT -08:00 US/Canada Pacific
Subject: Re: Wafer broken

Thanks for the information. 
My wafers might be weakened by the trenches significantly. At least they can gather the strain. I didn't see where the fracture occurred, since the wafers were broken into small pieces. 
I agree the the pick up process is rather violent. It's mostly likely that's when it breaks the wafers. It would be nice if it can be turned off.


chong

----- Original Message -----
From: "Christopher McGuinness" <cmcg at snf.stanford.edu>
To: "Chong Xie" <chongxie at stanford.edu>
Cc: "SangBum Kim" <kimsangb at stanford.edu>, cmp at snf.stanford.edu
Sent: Friday, April 10, 2009 1:59:29 PM GMT -08:00 US/Canada Pacific
Subject: Re: Wafer broken

This is strange. I've used the CMP many times and have never had a wafer break on me. The only things I can think that would cause this discrepancy are: 
1) We load the wafers differently. I don't see how this can be the case though, as it is pretty straightforward... 
2) The tool parameters are different. Again, I run the tool at the standard settings, so unless something has degraded or changed since I last used it this would not seem to be the case. 
3) There is something about our wafers that is different. I polish wafers with a .6um thick layer of oxide, patterned and filled with a similarly thick layer of poly. There is very little surface topology on my wafers and I polish them down to the surface of the oxide, so basically it is planar the whole time. Could the 2um trench you mentioned be weakening the wafer? Does the fracture occur along the trench? 

I have also noticed that when the tool turns the vaccuum on again to pick up the wafer it is a rather violent thump. I wonder if this vaccuum turn on step could be looked at and adjusted to be less violent. Or just skipped entirely so you pick your wafer up off the pad.... 

Sorry to hear about all the destruction. That must be frustrating. 

-Chris 





On Fri, Apr 10, 2009 at 1:48 PM, Chong Xie < chongxie at stanford.edu > wrote: 


Thanks a lot! That's very likely the case. I will try turning off the vacuum next time. 

chong 



----- Original Message ----- 
From: "SangBum Kim" < kimsangb at stanford.edu > 
To: "Chong Xie" < chongxie at stanford.edu > 
Cc: cmp at snf.stanford.edu 
Sent: Friday, April 10, 2009 12:33:34 PM GMT -08:00 US/Canada Pacific 
Subject: Re: Wafer broken 

Hi Chong, 

Do you know during which CMP step your wafer gets broken? In my case, most of my broken wafers actaully got broken when it is picked back up again by the head after it is done polishing. If that is the case for you, you can avoid that by turning off the vacuum pump during your polish so that the wafer is not picked up. 

Thanks, 
SangBum 

----- Original Message ----- 
From: "Chong Xie" < chongxie at stanford.edu > 
To: cmp at snf.stanford.edu 
Sent: Friday, April 10, 2009 10:58:58 AM GMT -08:00 US/Canada Pacific 
Subject: Wafer broken 

Dear CMP users, 

The CMP broke most of my wafers (3 out of 4) last time I use it. Chucking was all right, but the wafers were broken during polishing. I was just using the normal setting: 

Platen: 50RPM 
Head1: 30RPM 
Head2: 60RPM 

Polishing pressure: 200gm/cm2 
Retainer pressure: 300gm/cm2 

My wafers have 2 um deep trenches on them. I thought that might be the reason, but the same conditions worked for me all right before. 
I was wondering whether I could tune any of these parameters. Dose anyone have similar problems before? 
Any input is appreciated, and thanks in advance. 


chong 



-- 
Christopher McGuinness 
SLAC - Advanced Accelerator Research Department 
Stanford University - Applied Physics Department 
www.christophermcguinness.com 



More information about the cmp mailing list