Instantaneous Fluid Film Imaging in Chemical Mechanical Planarization

1
Instantaneous Fluid Film Imaging in Chemical
Mechanical Planarization

• Daniel Apone, Caprice Gray, Chris Rogers, Vincent
  P. Manno, Chris Barns, Mansour Moinpour, Sriram
  Anjur, Ara Philipossian

2
Motivation

• Microelectronic devices continue to decrease in
  size current features are routinely smaller than
  100nm
• The semiconductor industry requires a deeper
  understanding of the physical processes involved
  in CMP to help attain smoother surfaces
• Using Dual Emission Laser Induced Fluorescence
  (DELIF) we can measure instantaneous fluid film
  thicknesses (and temperatures) during a polishing
  run
• Here we look at how the pad conforms to features
  on a wafer

3
Polishing Setup

• Struers RotoPol-31 table top polisher
• Polisher sits atop a force transducer table
  capable of measuring down and shear forces during
  a polish

4
Optical Setup

• Evolution VF 12 bit digital cameras
• Region of Interrogation 3 cm across on pad
• Second ROI 3mm on pad

• 355 nm Nd-YAG Laser provides excitation light
• Laser Pulse Length 6ns

5
Dual Emission Laser Induced Fluorescence

• Calcein in slurry solution
• UV light excites Pads natural fluorescence
• Pads emission excites Calcein
• Each emission is captured by a camera
• Taking the ratio of the two emissions normalizes
  the image by initial excitation intensity

6
(No Transcript)
7
Experimental Parameters

• Freudenberg FX9 Pad
• Wafer Platen Rotation 30 rpm
• Relative Velocity 0.34 m/s
• Downforce 1.8 PSI
• Slurry
• Flow Rate 50 cc/min
• 91 dilution
• 0.5 g/L Calcein

8
Results

• Images are 3 cm viewing area on pad
• Air bubbles contained in a wave of slurry
• Striations made by conditioner
• Small circles are shadows of dried slurry on top
  of wafer

9
Previous Work

• Film thickness are measured from the wafer
  surface down to some mean height within the pad

• Film thickness increases as pad speed increases
• Inverse relationship for downforce and thickness

10
Searching for Contact.

• Images are 3 mm viewing area on pad can see
  individual asperities
• Dark areas have less fluid, indicate peaks
• Bright areas are holes in pad, more fluid there
• 10psi static image, to make sure contact was
  occurring
• Contact points seem to be few and far between

11
Pad Topology
12
Conclusion

• Pad topology seems to be the governing factor as
  to whether or not Pad/Wafer contact is occurring.
• Wafer seems to be supported by only a few peaks
  at any given time, the vast majority of
  asperities do not reach up to the wafer.

13
Future Work

• Investigate much larger region, to view multiple
  contact points in one image
• Ability to resolve individual asperities is
  necessary to determine if contact is occurring
• Correlate applied pressure with amount of
  contact?
• Correlate amount of contact with changes in
  friction data?

14
The End

• Acknowledgements
• Intel
• Cabot Microelectronics
• University of Arizona
• Questions?