IBM 8: Spin Coating

1
IBM 8 Spin Coating

• Greg Burtt
• Greg Hewitt
• David Valente
• Contact Engineer Kevin Remillard
• Faculty Mentor Jeff Marshall

2
Original Cleaning Disk
3
Bowl Set
4
Final Design Overview

• Solution 1 Modify the Cleaning Disk
• It has been determined through mathematical
  analysis and a physical inspection that the
  current cleaning disk has design flaws
• These include but may not be limited to the
  following
• Collection area for solvent is wide and flat, it
  appears that a significant portion does not flow
  through the disk
• A large majority of solvent sprays only
  horizontally, causing the bowl set to collect
  resist above and below this level
• The exit area for solvent is too small, an
  appreciable amount of solvent appears to overflow
  out of the cleaning disk
• Solution 2 Coat the Bowl Set
• If the bowl set were to have a coating that
  defied the buildup of resist, or was readily
  expurgated, it would aid the cleaning process and
  compliment Solution 1

5
Explanation of the Solutions

• Solution 1
• A preliminary prototype cleaning disk has been
  decided upon through the use of a Pugh chart
• This prototype increases solvent flow to the
  entire bowl set
• Above the horizontal, at the horizontal, below
  the horizontal, and the bottom bowl, around the
  knife edge
• The solvent exit velocity (and therefore mass
  flow rate) has been increased through the use of
  parabolic channels that will increase the height
  of fluid in the spinning disk
• A radius at the point of impact of the solvent on
  the disk will cause more solvent to be collected
  and less to splash back off, a smaller opening
  will also be employed to catch as much spray as
  possible
• Solution 2
• The company makes a coating called tt-Kote
  specifically designed for this application, we
  plan to apply it to a second set of bowls.

6
Pugh Chart for Disk Design
7
Prototype Disk
8
Testing and Refining the SolutionsPart 1

• A test stand will be built to replicate the
  equipment used at IBM
• This will consist, in part, of an acrylic glass
  case
• Inside the case there will be a motor to spin the
  cleaning disk, and to spin a wafer when
  performing tests to better determine where resist
  builds up on the bowls
• There will be gravity fed nozzles to spray the
  solvent into the cleaning disk from the back side
• A set of bowls will also be contained within, but
  will be removable
• A blower will be connected to the bowl sets to
  create the correct airflow

9
Test Stand
10
Testing and Refining the SolutionsPart 2

• After creation of the test stand, a number of
  tests will be performed
• The current cleaning disk will be tested to
  compare with the calculations, as a check of
  their accuracy
• Simulated photoresist (magenta inkjet printer
  ink) will be applied to the bowl sets, both
  coated and uncoated, using a spinning plastic
  disk in place of a silicon wafer
• This allows for an appropriate characterization
  of where resist builds up
• The prototype cleaning disk design will then be
  modified based on the results and an actual
  prototype will be manufactured for testing

11
Testing and Refining the SolutionsPart 3

• Both cleaning disks will be tested in both sets
  of bowls to determine two things
• If the prototype performs as expected, or what
  needs to be modified
• If the coated bowl outperforms the uncoated bowl
• Determine if the performance increase outweighs
  the cost of the coating
• Refine the prototype design and repeat

12
Pert Chart
Manufacture Prototype 02/22
Receive Parts 01/14
Assemble Test Stand 01/21
Test Current Cleaning Disk 02/04
Refine Prototype Design 02/08
Results 04/30
Ship Bowl for Coating 01/21
Bowl/Disk Testing 03/07
Receive Coated Bowl set 04/01
Coated Bowl/Disk Testing
Manufacture Final Prototype 04/01
Final Testing 04/09
Return On Investment 12/21
This can be postponed if necessary
Critical Path - - - - - Non-Crucial
Path Can be postponed if coated bowl set
arrives late
13
Work Breakdown Task List
Task Member(s) Involved Dates Estimated Hours Cost Deliverables
Return on Investment GB, GH, DV 12/07- 12/21 4 0.00 ROI Spreadsheet
Part Receiving and Inventory GB 01/14-01/21 2 1,650.00 Inventory Database
Test Stand Assembly GB, GH, DV 01/14-01/21 10 50.00 Test Stand
Ship bowl for Coating GH, Kevin 01/14-??? 0.5 800.00 Coated Bowl
Write Test Plan GH 01/14-01/21 4 0.00 Test Plan
Characterizing Original Disk GB, GH, DV 01/21-02/04 10 0.00 Baseline Dataset
Refine Initial Prototype Disks GB, GH, DV 01/21-02/08 5 0.00 Prototype Disk Design
Manufacturing Prototype DV 02/08-02/22 5 200.00 Prototype Disk
Modify Test Plan GB, GH, DV 02/08-02/22 1 0.00 Prototype Test Plan
Prototype Disk Bowl Testing GB, GH, DV 02/22-03/07 10 0.00 Prototype Dataset
Refine Prototype Design GB, GH, DV 02/22-03/07 5 0.00 Final Prototype Design
Manufacturing Final Prototype DV 03/07-04/01 5 200.00 Final Prototype Disk
Final Testing GB, GH, DV 04/01-04/09 10 0.00 Final Dataset
Results Processing GB, GH, DV 04/01-04/30 20 0.00 Final Results
91.5 2,900.00
Coated bowl data included if received within timeframe. Coated bowl data included if received within timeframe.
14
Charter

• By April 30th, 2008 a final cleaning disk
  prototype, cleaning recipe, and bowl coating
  decision will be made that will reduce the
  frequency with which the bowls are removed for
  cleaning. Initial testing will be performed to
  observe cleaning solvent behavior and collect
  empirical data on the performance of the current
  cleaning disk, and compare performance of the
  current bowl with a bowl coated with Nx-Edge
  tt-Kote. After initial testing, the disk design
  will be refined and a prototype will be
  manufactured and tested in both the coated and
  uncoated bowlsets. Modifications and testing will
  continue to be performed on the prototype disk
  until a satisfactory final design is engineered.
  The results of the testing, the final cleaning
  disk design, cleaning recipe, and bowl coating
  decision will be submitted to IBM.