CMP Pressure Distribution Study Group

Final Project UpdateBy

Dave BullenAlia Koch

Alicia Scarfo7/30/1999

Overview

• Previous work

• Photos: The Jumbo 1000 in action

• Data collection software redesign

• Final static and dynamic pressure data

• Numerical work on pad deformation

Previous Work

• Manometer Testing– Response was to slow– Could not measure a wide enough range

• Pressure Gage Testing– Adequate response time– Still could not read a wide enough range– Could not be used with a rotating wafer

The Jumbo 1000 in Action

The Jumbo 1000 in Action

The Jumbo 1000 in Action

Data Acquisition

• LabVIEW helped because: – Real Time Data– Very Accurate Data from Transducer– Easier to Read Data– Other Calculations made in Real Time

The Program

• We were able to control:– Scan Rate– Scans per Channel– Total Scan Time– Total Number of Scans

Our Front Panel

Wafer Static PressuresPressure vs. Position: All Taps, Static, Transducer

56789

10111213141516171819202122232425262728

0 30 60 90 120 150 180 210 240 270 300 330

Position (Degrees)

Pre

ss

ure

(p

sia

)

Run details:Date: 7/27/99-7/28/99Wafer: REU-1, StaticPad Speed: 120 RPMConditioning: Yes - 8 oscillations per minuteFluid: Water with Sodium Hydroxide - pH: 11.27 - 11.30Applied Pressure: 3 psiFlow separation: None observed

Tap 6

Tap 5

Tap 4

Tap 1

Tap 3

Tap 2

Atmospheric

Downforce

Wafer Static Pressures

• Pressure distribution is divided into two high pressure regions and two low pressure regions

• There is a wide variation in the pressures from point to point

• There is a vague increasing trend from rim to center

Wafer Dynamic PressuresPressure vs. Position, All Taps, Dynamic Averages

14

15

16

17

18

19

20

21

22

Position

Pre

ssu

re (p

sia)

Tap 1

Tap 2

Tap 3

Tap 4

Tap 5

Tap 6

Note: Tap locations are unknown. This plot is assembled so that all peaks line up. This may not actually be the case.

Wafer Dynamic Pressures

• Pressure distribution structure is similar to the static structure

• Less variation from point to point

• There is a strong increasing trend from rim to center

Accomplished Numerical Goals

• Wrote structural program– modeled pad changes using finite element

methods– wrote program in Fortran 77 compatible with

current Fidap fluid finite element program

Update

• Program now compiles!

• Future changes to structural FEM program– program would benefit from new mesh generator– observe effect of pad deformation on pressure

distribution under the wafer

• Other options– try to incorporate strain program supplied by

Professor Perlman

Project Accomplishments

• Determined the pressure distribution under a static and dynamic wafer via three methods

• Created a system that can monitor film pressure under a rotating wafer

• Used numerical methods to advance the understanding of the pad deformation phenomenon

Future Pressure Detector Work

• Manufacturer the rotating platform and glass wafer as specified.

• Add a position signal to the data stream.

• Upgrade the sliding contact system to transmit several channels.

• Redesign transducer/air purge system to further reduce the response time.

Thanks

Dewi Bramono

Dr. Jonathan Coppeta

Dr. Livia Racz

James Hoffman

Joe Lu

The whole TAMPL gang