microelectronic engineering rochester institute ... - people · manufacturing vs fabrication ......
Post on 17-Aug-2018
220 Views
Preview:
TRANSCRIPT
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 1© Nov ember 25, 2012 Dr. Lynn Fuller
MICROELECTRONIC ENGINEERINGROCHESTER INSTITUTE OF TECHNOLOGY
11-25-2012 lec_mfg.ppt
Manufacturing vs Fabrication
Dr. Lynn FullerWebpage: http://people.rit.edu/lffeee
Microelectronic EngineeringRochester Institute of Technology
82 Lomb Memorial DriveRochester, NY 14623-5604
Tel (585) 475-2035Fax (585) 475-5041
Email: Lynn.Fuller@rit.eduDepartment webpage: http://www.microe.rit.edu
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 2© Nov ember 25, 2012 Dr. Lynn Fuller
ADOBE PRESENTER
This PowerPoint module has been published using Adobe Presenter. Please click on the Notes tab in the left panel to read the instructors comments for each slide. Manually advance the slide by clicking on the play arrow or pressing the page down key.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 3© Nov ember 25, 2012 Dr. Lynn Fuller
OUTLINE
IntroductionFabrication vs ManufacturingManufacturing Early 1990’s, Today, Tomorrow
Semiconductor Manufacturing Includes:Factory Modeling: Logistics (Scheduling), Ramp UpFactory Floor Control, WIP TrackingCycle Time ManagementContamination Free Manufacturing, Yield, Metrology, TestProcess Integration, Evolution and Improvement,
Advanced Processes,Technology TransferTQM, CIM, SPC, 6 Sigma Manufacturing, DOE,
Statistical ThinkingCost Containment
Human ResourcesUndergraduate EducationUniversity IC Labs
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 4© Nov ember 25, 2012 Dr. Lynn Fuller
FABRICATION VS MANUFACTURING
FABRICATION involves the design and realization of a semiconductor device or circuit. The goal is achieved if one device or circuit is made to work. Research is centered on new technologies and materials, new, smaller and faster devices, novel circuits, etc.
MANUFACTURING involves the realization of a large number of semiconductor devices or circuits. The goal is achieved if large numbers of circuits are made, at low cost (at a profit), with high yield and quick turn around time. Research is centered on manufacturing methodology, operations research, statistical process control, factory simulation, process integration, etc.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 5© Nov ember 25, 2012 Dr. Lynn Fuller
SEMICONDUCTOR MANUFACTURING IN 1991
NUMBERS FROM A MEDIUM SIZE SEMICONDUCTOR PLANT AT INTEL
WITH 5000 WAFER STARTS/WEEK, 0.7 µm PROCESS
5000 WAFERS 150 mm DIAMETER PER WEEK7 DAYS/WEEK, 24 HOURS/DAY OPERATION300 EMPLOYEESCOST OF PLANT $500 MILLIONLIFE OF PLANT 5 YEARSWAFER LOT SIZE = 25COST OF PROCESSING ONE LOT = $25,000SALES VALUE OF ONE LOT = $100,000 ($2.8 MILLION/DAY)TIME IN MANUFACTURING = 40 DAYSDISTANCE TRAVELLED IN FAB = 6 MILES13 PEOPLE TO JUST MOVE, LOAD, UNLOAD MACHINESMACHINE LOAD AND UNLOAD CYCLES = 6.8 MILLION/YEAR20,000 MACHINE LOAD AND UNLOAD PER DAYWORK IN PROCESS = 15,000 WAFERSVALUE OF WORK IN PROCESS = $60 MILLION
SOURCE: E.SHAMAH MAY 1991, SRC PRESENTATION
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 6© Nov ember 25, 2012 Dr. Lynn Fuller
SEMICONDUCTOR MANUFACTURING IN 2000
PARTIAL CIMTUNNEL AND CHASE FACILITY DESIGNCASSETTE TO CASSETTELIMITED AUTOMATIC MATERIALS MOVERSOUTPUT PARAMETER SPCEXSITU METROLOGYBATCH PROCESSINGCLASS 10 TO 1 VLFLIMITED INTEGRATED PROCESS TOOLS90% FAB YIELD60% SORT YIELD40% EQUIPMENT UTILIZATION2X THEORETICAL CYCLE TIME0.5 DEFECTS PER CM2 AT 1 MICRON100-150 mm WAFER SIZE
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 7© Nov ember 25, 2012 Dr. Lynn Fuller
SEMICONDUCTOR MANUFACTURING in 2010
200 mm Wafer Diameter (8”)
Ballroom Facility Design
8,000 wafers per week
0.18 µm5 levels metal, CMP
Cost of Plant $2 Billion
Equipment Reliability, Uptime Must be Maximized
100% Total Preventative Maintenance
Inter-Bay and Intra-Bay AutomationAdvanced WIP Tracking to Eliminate Queues
Automated Tools and Recipe Handling
Extend the Useful Life of Tools
Hug the 100% Loading Capacity
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 8© Nov ember 25, 2012 Dr. Lynn Fuller
TOMORROW’S SEMICONDUCTOR MANUFACTURING TODAY
I.C. FACTORY COST $3 BILLION
BALLROOM, MINI-ENVIRONMENTS
300 mm WAFER DIAMETER (12”)
TOTAL AUTOMATION
GIGABIT DRAMS
150 MILLION TRANSISTOR I.C.’S
LESS THAN 0.045 MICRON FEATURES
LESS THAN 0.01 DEFECTS/CM2
2000 MHZ
2 BILLION INSTRUCTIONS/SEC
1.5 VOLT
1 INCH BY 1 INCH CHIPS
400 LEADS
25 WATTS/CHIP
NEW DESIGN TOOLS FOR
BILLION TRANSISTOR CIRCUITS
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 9© Nov ember 25, 2012 Dr. Lynn Fuller
CLEANROOM DESIGN
Ballroom Design
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 10© Nov ember 25, 2012 Dr. Lynn Fuller
SEMICONDUCTOR MANUFACTURING INCLUDES
Factory Modeling: Logistics (Scheduling), Ramp Up, Factory Floor Control, WIP Tracking,
Cycle Time Management, Time to MarketContamination Free Manufacturing, Yield, Metrology, TestProcess Integration, Evolution and Improvement,
Advanced Processes, Technology TransferTQM, CIM, SPC, 6 Sigma Manufacturing, DOEStatistical ThinkingCost Containment: Cost of Ownership, Resource Modeling,
Activity Based Costing, Elimination of Non Value Added Activities
Human Resources
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 11© Nov ember 25, 2012 Dr. Lynn Fuller
FACTORY MODELING
Server1
Server4
Server3
Server2
End P2
Start P2
Start P1
End P1
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 12© Nov ember 25, 2012 Dr. Lynn Fuller
FACTORY MODELING
IFR_300mm_trailer.wmv
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 13© Nov ember 25, 2012 Dr. Lynn Fuller
FACTORY PROCESS FLOW
32
1
15
98
7
6
4
5
0
11 10
14
13
1617
RIT p-well CMOSprocess flow, 60 steps,9 photo steps, 7 oxidesteps, 6 implants
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 14© Nov ember 25, 2012 Dr. Lynn Fuller
WIP TRACKING
Logistics, Ramp Up, Factory Floor Control, WIP Tracking
Thousands of wafers say 25,000 in the factory, what to do next.300 products each with different mask sets for a total of 6000 masks to keep track of. 250,000 turns per day. 8000 wafer starts per week.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 15© Nov ember 25, 2012 Dr. Lynn Fuller
LOGISTICS
LOT SELECTION RULES
Do Photo first
Do Oldest Lot Next
Separate Lots Current Step
Match Skill Level
Use Equipment that is Up
Apply Lot
Selection
Rules
Check
Equipment
Status
In Queue?
AccessMESALot Status
On Hold?
On Hold?
Prelininary
Quality
Check
Find
Wafers
Do Work
Follow MESA
Instructions Exactly
Do Move-In
Start Run
TimerPass ?
FINAL QUALITY CHECK
Count Wafers
Check Picture Log Book
Think
Do Results Make Sense?
Contact Person
Determine What
To Do Next
Final
Quality
Check
Yes
YesSee Lab Instructor
Yes
No
No
No
No
INITIAL QUALITY CHECK
Count Wafers
Check Picture Log Book
Think
Refer to Previous Process Step
Check MESA Move-Out Comments
START
Mesa
History
Who Did
Move-In
Find Queue Status
Step Number
Current Operation
Next Operation
Quantity
See Lab Instructor
Pass ?Clean Up
Return Wafers
Return Masks
Stop Run Timer
Move Out
Record Data
ENDNo
See Lab Instructor
Yes
Continue
A
Continue
A
Yes
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 16© Nov ember 25, 2012 Dr. Lynn Fuller
CYCLE TIME MANAGEMENT
CYCLE TIME - the time it takes to process wafers from start to finish. Various cycle times can be calculated depending on the exact definition. Usually cycle time is the number of calendar days to process a lot from start to ship. Other variations include single wafer cycle times, cycle time based on work days rather than calendar, etc.
BASELINE CYCLE TIME (work days), (BSWCT and BWLCT) - this is the cycle time at the start of a cycle time improvement program. At that point in time a CIM system data base query is done to find the cycle time for each process flow (PMOS, NMOS, CMOS, EEPROM, etc.) This is used as the reference point for measuring cycle time improvement.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 17© Nov ember 25, 2012 Dr. Lynn Fuller
CYCLE TIME IMPROVEMENTC
ycle
time,
X f
acto
r(t
imes
theo
retic
al c
ycle
time)
161514131211109876543210
Ap
r
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Fab 2 Cycletime
standarddeviation
Ave
range
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 18© Nov ember 25, 2012 Dr. Lynn Fuller
DEFECT REDUCTION
Particle Count in Vicinity of Photoresist Spinner
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 19© Nov ember 25, 2012 Dr. Lynn Fuller
DEFECT DENSITY AND DIE YIELD
YIELD = e -AD
Where: A is the chip area (cm2) and D is the density of defects (# / cm2)
EXAMPLE: Chip Area is 1 cm2 and Defect density is 1/cm2
YIELD = e -1 = 37%
Jan98
Jan97
Jan96
Jan95
0.1
1
100
10 Joe Juran Methodology Continuous Measurable
ImprovementYield
Time
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 20© Nov ember 25, 2012 Dr. Lynn Fuller
YIELD AND MANUFACTURING
DIE YIELD = NUMBER OF WORKING CHIPS/TOTAL NUMBER OF CHIPS
WAFER YIELD = [WAFER YIELD / STEP](NUMBER OF STEPS)
Example: YIELD = 98%(100) = 13%
Example: DIE YIELD = 52/61 = 85%
COST OF DIE YIELD LOSS
= 5000 wafers/wk x 52 wk/yr x $5/chip x 9chips/wafer = $1.2 million/yr
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 21© Nov ember 25, 2012 Dr. Lynn Fuller
PROCESS INTEGRATION
PROCESS IS: A VARIETY OF SEQUENTIAL STEPS WHICH RESULTS IN HUNDREDS OF THOUSANDS OF TRANSISTORS BEING MADE AT THE SAME TIME ON EACH CHIP
UNIT PROCESSES ARE:DEPOSITION - CVD, LPCVC, PECVD, PVDSURFACE ALTERING - DIFFUSION, OXIDATION, ION IMPLANTATIONPHOTOLITHOGRAPHY - G-LINE, I-LINE, EXCIMER LASER, X-RAYETCHING - WET CHEMICAL ETCHING, PLASMA ETCHING, RIECLEANING - RCA, MODIFIED RCA
PROCESS INTEGRATION: IS THE SEQUENCING THE UNIT PROCESSES INTO A SERIES OF STEPS THAT GIVES THE DESIRED RESULT
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 22© Nov ember 25, 2012 Dr. Lynn Fuller
EVOLUTION OF MANUFACTURABILITY
MATURE
PROCESS
PUSH FOR NEW
PROCESS
TECHNOLOGY
PROCESS
CAPABILITY
IMPROVEMENT
NEW
PRODUCTSLOW
MANUFACTURABILITY
HIGH
MANUFACTURABILITY
NEED FASTER CHIPS
MORE FUNCTIONS
INTRODUCTION TO
MANUFACTURING
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 23© Nov ember 25, 2012 Dr. Lynn Fuller
TECHNOLOGY TRANSFER
Technology is Transferred by PeopleTrainingExchanges
Copy ExactlyAttention to DetailDocumentation
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 24© Nov ember 25, 2012 Dr. Lynn Fuller
TQM - TOTAL QUALITY MANAGEMENT
“you are not paid for coming to work. You are paid to make the product
better”
TQM is a way to run a company that focuses on continually improving how you do what you do in order to satisfy all customer needs. TQM combines management methods and statistical tools in one package and gives all members of an organization a common goal.
Quality is defined as “conformance to customer requirements”, lack of defects and the companies own standards of final product.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 25© Nov ember 25, 2012 Dr. Lynn Fuller
CIM - COMPUTER INTEGRATED MANUFACTURING
CIM - Concept in which computer software and hardware is integrated throughout a manufacturing facility to provide integration among functions such as engineering and research, production planning, plant operations, shipping, receiving, business management, marketing, everything (including CAD and CAM)
CAD - Computer Aided Design, software and hardware tools needed to design the product including, circuit simulators, layout editors, process simulators, and more,
CAM - Computer Aided Manufacturing - software and hardware tools needed for work in process tracking, statistical process control, facilities monitoring, robotics, artificial intelligence, expert systems, and more.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 26© Nov ember 25, 2012 Dr. Lynn Fuller
COMPUTER AUTOMATED SEMICONDUCTOR MANUFACTURING
Process EngineeringCAD, CAM, SPC START
Operation 2
Operation 1
Final Test
Operation N
SHIP
Data Collection,Lot Control,Scheduling,SPC, Rework
Facilities, Equipment,SuppliesPeople, Training, Technical Support
Expert System
Robotics
Evaluate
Feed Forward
Feed Back
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 27© Nov ember 25, 2012 Dr. Lynn Fuller
AUTOMATION
Material Handlingand Control Automation
Workflow Automation:People monitor
and manage exceptions
Equipment Automation:Hands-off
processing
Process Automation:Quality control
based on machine data
Goal:
All execution and control
fully automated
MaterialBuffer
OverheadTransport
Process&MetrologyEquipmentComputer Integration
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 28© Nov ember 25, 2012 Dr. Lynn Fuller
FULLY AUTOMATED 250,000 TIMES PER DAY
1.Processing finished2.Data to SPC and R2R uploaded3.Transport to next area started
4. Lot for next tool reserved 5. Carrier transport to tool
6. Process parameter determined
7. Processing started
Metrology
Etch
SPC: Statistical Process controlR2R: Run-to-run
Lithography
Metrology
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 29© Nov ember 25, 2012 Dr. Lynn Fuller
RUN TO RUN CONTROL
Lithography Dry Etch
Feed forward
Run-to-Runcontrol
Small fluctuations
of processingCompensation
of variationsFeed back
Run-to-Runcontrol
Measurement Parameter
adjustment
Parameter
adjustment
Goal: Automatically minimize process variations
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 30© Nov ember 25, 2012 Dr. Lynn Fuller
RUN TO RUN CONTROL
-350
-320
-290
-260
-230
-200
-170
-140
-110
-80
-50
-20
10
40
70
100
130
160
190
220
250
280
310
340
No
rma
lize
d V
alu
e (
De
lta
Ta
rge
t in
An
gst
rom
s)
10
/23
/20
05
10
/26
/20
05
10
/29
/20
05
11
/01
/20
05
11
/04
/20
05
11
/07
/20
05
11
/10
/20
05
11
/13
/20
05
11
/16
/20
05
11
/19
/20
05
11
/22
/20
05
11
/25
/20
05
11
/28
/20
05
12
/01
/20
05
12
/04
/20
05
12
/07
/20
05
12
/10
/20
05
12
/13
/20
05
12
/16
/20
05
12
/19
/20
05
12
/22
/20
05
12
/25
/20
05
12
/28
/20
05
12
/31
/20
05
01
/03
/20
06
01
/06
/20
06
01
/09
/20
06
01
/12
/20
06
01
/15
/20
06
01
/18
/20
06
01
/21
/20
06
01
/24
/20
06
01
/27
/20
06
01
/30
/20
06
02
/02
/20
06
02
/05
/20
06
02
/08
/20
06
02
/11
/20
06
02
/14
/20
06
02
/17
/20
06
02
/20
/20
06
02
/23
/20
06
02
/26
/20
06
03
/01
/20
06
03
/04
/20
06
03
/07
/20
06
03
/10
/20
06
03
/13
/20
06
03
/16
/20
06
03
/19
/20
06
03
/22
/20
06
03
/25
/20
06
03
/28
/20
06
03
/31
/20
06
04
/03
/20
06
04
/06
/20
06
04
/09
/20
06
04
/12
/20
06
04
/15
/20
06
04
/18
/20
06
04
/21
/20
06
04
/24
/20
06
04
/27
/20
06
04
/30
/20
06
05
/03
/20
06
05
/06
/20
06
05
/09
/20
06
05
/12
/20
06
05
/15
/20
06
05
/18
/20
06
05
/21
/20
06
SAMPLE_DATE
B
D
SUBUNIT
Without R2R
Man
ual
up
dat
es
usi
ng
R2R
se
ttin
gs
With Automated R2R
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 31© Nov ember 25, 2012 Dr. Lynn Fuller
SPC - STATISTICAL PROCESS CONTROL
§ CIM system integrated with SPC software - Quality Analyst
§ operators review SPC charts before processing
§ process adjustments can be made if necessary
§ SPC alarms and actions if process violates SPC rules:
§ send notice to specific users
§ prevent further processing of job, operation or tool
§ Corrective Actions
§ List of actions (flow chart) if SPC rules are violated
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 32© Nov ember 25, 2012 Dr. Lynn Fuller
STATISTICAL PROCESS CONTROL (SPC)
0
1000
2000
3000
4000
5000
6000
0 50
Oxi
de T
hick
ness
, Å
Date
Jan
1997
Sep
1997
June
199
7
Mar
199
7
USL
LSL
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 33© Nov ember 25, 2012 Dr. Lynn Fuller
STATISTICAL THINKING
LITHOGRAPHIC ETCH CELL FORMATION
Overall VariationNo Cells
Cell 2Cell 1
Cell 3
Coat 1
Expose 1Coat 2
Coat 3Expose 2
Dev 1
Dev 2
Dev 3
Etch 1
Etch 2
36 Possible Pathswith cell formation
3 Possible Pathsand less variation
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 34© Nov ember 25, 2012 Dr. Lynn Fuller
SIX SIGMA CONCEPTS
03σσσσ3σσσσ
LSL USL
Mean
03σσσσ3σσσσ
LSL
Target Mean
USL
03σσσσ3σσσσ
LSL USL
Mean
Process Variationwithin specification
limits
Process Variationlarger than
specification limits
Process Meanoff target
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 35© Nov ember 25, 2012 Dr. Lynn Fuller
SIX-SIGMA CONCEPTS
Reduce Variation - locate 6σσσσ points within USL and LSL & Position Mean at Target
Goal of Cp=2.0, k=0, Cpk=2.0 to give less than 3.4 ppm defects
Cp = (USL - LSL)/6 σ, σ, σ, σ,
K = [Tget - µ]/(USL - LSL)/2,
Cpk = Smallest of: {(µ-LSL)/ 3 σσσσ;(USL-µ)/ 3 σσσσ }
Target
0 3σσσσ 6σσσσ3σσσσ6σσσσ
Mean, µ
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 36© Nov ember 25, 2012 Dr. Lynn Fuller
DESIGN OF EXPERIMENTS (DOE)
Power, watts40 50 60
100
200
300
Pre
ssur
e, m
Torr
Central Composite Designs
What can you measure?linear effectsinteractionsquadratic effects
# factors # runs2 93 154 255 43
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 37© Nov ember 25, 2012 Dr. Lynn Fuller
DOE, RESPONSE SURFACE ANALYSIS
Model Equation: F = 5X2 - 1.5Y - 4XY + 12
Surface Plot Contour Plot
1
0
1 0.5 0 0.5 11
0.5
0
0.5
1
0.70.6
0.55
0.5
0.5
0.45
0.45 0.45
0.45
0.4
0.4
0.4
0.4
0.35
0.35
0.35
0.35
0.3
0.3
0.3
0.3
0.3
0.25
0.25
0.25
0.25
0.25
0.2
0.2
0.20.15
0.15
0.15 0.10.1
Tilt = 10° Rotation = 30°
(normalized response, X & Y in design units)
XY
F
X
Y
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 38© Nov ember 25, 2012 Dr. Lynn Fuller
COST OF OWNERSHIP (COO)
$COO =L x TPT x Y(THP) x U
$F + $V + $Y
Where:
$COO is Cost of Ownership in $/wafer
$F is Fixed Cost
$V is Variable CostL is Equipment Life
TPT is Throughput rate
Y(TPT) is Ideal Throughput Yiels
U is Utilization
$Y is Cost of Yield Loss
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 39© Nov ember 25, 2012 Dr. Lynn Fuller
INDUSTRY NEED FOR HUMAN RESOURCES
The semiconductor industry is forecasted to grow even larger. Worldwide industry sales are projected to reach $400 billion by the year 2010. Most projections indicate that the semiconductor industry will need about 40,000 more skilled operators and technicians during the next five years and 8,000 semiconductor manufacturing engineers (Process Engineers, Product Engineers, Device Engineers, Defect Reduction and Yield Enhancement, Test Engineers, Reliability, Process Integration ) in the next five years.
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 40© Nov ember 25, 2012 Dr. Lynn Fuller
TYPICAL UNDERGRADUATE MICROELECTRONICS EXPERIENCE AT A UNIVERSITY
VLSI Design (or Digital Systems)Analog IC Design (Electronics)
IC TechnologySemiconductor Device Fabrication Laboratory
Device Physics
All have very little Manufacturing ContentApproximately 1000 students/year
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 41© Nov ember 25, 2012 Dr. Lynn Fuller
DESIGN
VLSI DesignVHDLAnalog IC DesignCircuit SimulationLayoutDesign AutomationDesign Rule Checking
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 42© Nov ember 25, 2012 Dr. Lynn Fuller
IC TECHNOLOGY
OxidationDiffusionCVD/LPCVD/PECVDPlasma EtchRapid Thermal AnnealPhysical Vapor DepositionIon ImplantLithographyTechnology Modeling
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 43© Nov ember 25, 2012 Dr. Lynn Fuller
SEMICONDUCTOR DEVICES
ElectronicsDiodes, TransistorsAnalog Building Blocks
Physics of SemiconductorsMOS StructuresTransistors
Microelectromechanical DevicesSensors & Actuators
Semiconductor DevicesCCD’s, CID’s
Non-Silicon DevicesGaAs, III-V, II-VIFlat Panel DisplaysLiquid Crystal Devices
Device Simulation
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 44© Nov ember 25, 2012 Dr. Lynn Fuller
MICROLITHOGRAPHY
Exposure Tools, g-line, I-line, 248nm, 193nm, e-beamCoat and Develop ToolsResist Materials
Positive Novalac ResistsNegative Chemically AmplifiedContrast EnhancementDyed, ARCMultilayerTop Surface ImagingDUV, EUV, X-ray
ModelingPhase Shifting MasksDUV, EUV
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 45© Nov ember 25, 2012 Dr. Lynn Fuller
MATERIALS SCIENCE
Surface Analysis
SEM, TEM (EDAX)AUGER, SIMSXPS, ESCA
Materials Processing
Thin Films (metals, insulators)CVD (LPCVD, PECVD, etc.)Plasma EtchingRapid Thermal Processing
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 46© Nov ember 25, 2012 Dr. Lynn Fuller
SEMICONDUCTOR MANUFACTURING
Operations Research - Factory Floor Simulation, WIPTracking, Cycle Time Management, Materials Resource Planning, Scheduling, Productive Maintenance
Manufacturing - Process Engineering, Statistical Process Control, Process Capability Analysis, TQM, CIM, Cycle Time, Defect Reduction and Yield Enhancement
Statistical Process Control - DOE, Statistical Thinking, Time Series Analysis
Computer Automation - CAD, CAM, CIM, SECS I,II, Robotics, AI, Expert Systems
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 47© Nov ember 25, 2012 Dr. Lynn Fuller
UNIVERSITY IC LABORATORY
Most university IC labs exist to support research. Some are also used as educational IC labs. Sometimes separate IC labs are maintained for education.
The educational IC labs in universities exist to support courses in semiconductor technology, semiconductor devices, semiconductor manufacturing, microlithography and/or materials science.
Around 40 universities have IC lab facilities
About 15 universities have large complete facilities (> 10,000 sq.ft.)
Community Colleges are also starting programs to support the semiconductor industry and they are building educational facilities
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 48© Nov ember 25, 2012 Dr. Lynn Fuller
EDUCATIONAL LAB FACILITIES
BASIC
3-6 Tubes FurnaceResist Spinner
Bake oven or Hot PlateContact Printer/Aligner
Wet Etch HoodEvaporator
Optical Microscope4pt. Probe
Groove and StainSUPREM II
Manual ProberHP4145
CMOS CAPABLE
Basic FacilityPlus All Below:
LPCVD, Poly and NitrideWafer Coat/Develop System
SteppersPlasma Etch or RIE
SputteringIon Implant
SEM, Nanospec, AlphastepSUPREM III, IV
PROLITH
ADVANCED
CMOS Capable FacilityPlus Some Of Below:
Maskmaking(e-beam or Optical)
CIM SystemEDAX, AUGER, SIMS
XPS, ESCAPackaging
Semi Automatic Probeand Test
MBE, MOCVD, RTP
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 49© Nov ember 25, 2012 Dr. Lynn Fuller
UNIVERSITIES
Arizona State UniversityBoise State University*Boston UniversityBrown UniversityCarnegie-Mellon UniversityCase Western Reserve UniversityColumbia UniversityCornell University*Duke UniversityFlorida Institute of TechnologyGeorge Washington UniversityHarvard University*James Madison UniversityMassachusetts Institute of Technology*New Mexico State UniversityNorth Carolina A&T UniversityNorth Carolina State University*Northwestern UniversityPennsylvania State UniversityPurdue University*
Rensselaer Polytechnic Institute*Rochester Institute of Technology*Rutgers UniversityStanford University*San Jose State UniversitySanta Clara State UniversityTufts UniversityUniversity of ArizonaUniversity of California at Berkeley*University of California at Santa
BarbaraUniversity of California at Los AngelesUniversity of CincinnatiUniversity of FlordiaUniversity of Illinois*University of LouisvilleUniversity of MarylandUniversity of MassachusettsUniversity of Michigan*University of Minnesota*University of Mississippi
University of NebraskaUniversity of OklahomaUniversity of PennsylvaniaUniversity of RochesterUniversity of South CarolinaUniversity of Texas at ArlingtonUniversity of Texas at Austin*University of UtahUniversity of VirginiaUniversity of WisconsinVirginia Commonwealth University*
These Universities have IC Fabrication Laboratory Facilities
* Large IC Fabrication Facilities
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 50© Nov ember 25, 2012 Dr. Lynn Fuller
TWO YEAR SCHOOLS
Central Arizona CollegeChandler/Gilbert Community CollegeGateway Community CollegeGlendale Community CollegeMesa Community CollegePima Community CollegeMission CollegeSna Jose City CollegeOrange Coast CollegePikes Peak Community CollegeAims Community CollegeValencia Community CollegeIdaho StateSouthern Maine Technical CollegeWorcester Technical InstituteAlbuquerque TVINorthern NM Community CollegeSan Juan College
Santa Fe Community CollegeLuna Vocational Technical InstituteSW Indian Polytechnic InstituteDona Ana Community CollegeUniversity of New Mexico Valencia CampusChemeketa Community CollegeOregon Institute of TechnologyPortland Community CollegeUmpqua Community CollegeLinn Benton Community CollegeAustin Community CollegeCollin County Community CollegeEastfield CollegeGrayson CollegeMountain View CollegeNorth Central Texas CollegeNorth Lake CollegeRichland CollegeTarrant County Junior College
Texas State Technical College-Harlingen
Texas State Technical College-Sweetwater
Texas State Technical College-Waco
Weatherford CollegeVermont Technical CollegeCentraliaPierce CollegeDallas County Community CollegeCollin County Community College
These Community Colleges are developing programs to Prepare students for semiconductor manufacturing
technology positions
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 51© Nov ember 25, 2012 Dr. Lynn Fuller
MICROELECTRONICS EDUCATION IN UNITED STATES UNIVERSITIES
Number of ABET Accredited EE Programs 250Students in EE Programs 100,000 = 25,000/yearStudents taking VLSI Design 5,000/yearStudents in Fabrication Courses 1,000/yearGraduate Students Studying Semiconductors 2,400=300/yearChemical Engineering Programs
with Microelectronics Courses 10Undergraduate Microelectronics Programs 10
RIT has the only ABET accredited BS program in
Microelectronic Engineering
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 52© Nov ember 25, 2012 Dr. Lynn Fuller
RIT PROGRAM
Started in the Fall of 1982ABET Accredited125 Undergraduate Students15 Masters Students5 Ph.D. Students5 year Required Co-op Program for UndergraduatesOver 1000 Graduates
Addresses all Aspects of Microelectronics Manufacturing Starting in the First Year
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 53© Nov ember 25, 2012 Dr. Lynn Fuller
REFERENCES
1. ITRS International Technology Roadmap for Semiconductors, http://www.itrs.net/
2. Semiconductor Subway,
http://www-mtl.mit.edu/semisubway/
3. Semiconductor International Magazine,
http://www.e-insite.net/semiconductor/
4. Solid State Technology Magazine, http://sst.pennnet.com/home.cfm
5. EE Times, http://www.eetimes.com/
Rochester Institute of Technology
Microelectronic Engineering
Manufacturing vs Fabrication
Page 54© Nov ember 25, 2012 Dr. Lynn Fuller
HOMEWORK - MANUFACTURING
1. Look at all the references listed on the previous page.
top related