Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
MicroForce™ Probing for Devices MicroForce™ Probing for Devices with Lowwith Low--k ILD Materialsk ILD Materials
Rod Martens, Ph.D., FormFactor, Inc.Rod Martens, Ph.D., FormFactor, Inc.
Abdel Abdelrahman, Intel CorporationAbdel Abdelrahman, Intel Corporation
Steve Martinez, Tokyo Electron LimitedSteve Martinez, Tokyo Electron Limited
Southwest Test Conference, Long Beach, CASouthwest Test Conference, Long Beach, CA
June, 2003June, 2003
2SouthWest Test Conference June, 2003
ContentsContents
Introduction to MicroForce™ Probing TechnologyIntroduction to MicroForce™ Probing Technology–– Industry Requirements for Probing Flip Chip Bump Devices Industry Requirements for Probing Flip Chip Bump Devices
–– Probe Card ArchitectureProbe Card Architecture
–– MicroForce Probing DefinitionMicroForce Probing Definition
FormFactor Lab DevelopmentFormFactor Lab Development–– Single Probe Testing of Force and CresSingle Probe Testing of Force and Cres
Sort Floor ValidationSort Floor Validation–– Force MeasurementsForce Measurements
–– Probe Mark ComparisonsProbe Mark Comparisons
–– Cres and Probe Conditions After Aging TestCres and Probe Conditions After Aging Test
ConclusionsConclusions
3SouthWest Test Conference June, 2003
Industry Requirements for Probing Industry Requirements for Probing Advanced Flip Chip Bumped DevicesAdvanced Flip Chip Bumped Devices
Lower Force ProbingLower Force Probing–– LowLow--k ILD materials are more fragilek ILD materials are more fragile
–– Low probe force desired to minimize damage to flip chip bump Low probe force desired to minimize damage to flip chip bump and ILD materialsand ILD materials
Low and Consistent Contact ResistanceLow and Consistent Contact Resistance–– High frequency and high power applications require tighter Cres High frequency and high power applications require tighter Cres
controlcontrol
–– Applications require higher current to pass through probe Applications require higher current to pass through probe system without much resistancesystem without much resistance
Finer PitchFiner Pitch–– Roadmap to 100Roadmap to 100µµm pitch requires revolutionary probing m pitch requires revolutionary probing
approachapproach
4SouthWest Test Conference June, 2003
FormFactor BladeRunner™ Probe Cards FormFactor BladeRunner™ Probe Cards (For Probing Bumped Wafers)(For Probing Bumped Wafers)
5SouthWest Test Conference June, 2003
FormFactor Probe Card ArchitectureFormFactor Probe Card ArchitectureControlled Impedance Tester Interface
PCB Assembly
InterposerWith MicroSpring
Contact
Space Transformer:
MicroSpring™ Contact for Probing Flip Chip Bumps
6SouthWest Test Conference June, 2003
MicroForceMicroForce™™ Probing DefinitionProbing Definition
What is MicroForce™ probing?What is MicroForce™ probing?–– It is a coordinated XIt is a coordinated X--YY--Z probing motionZ probing motion
–– Developed to satisfy a number of stringent requirements for Developed to satisfy a number of stringent requirements for probing devices with lowprobing devices with low--k ILD materialsk ILD materials
–– Breakthrough results with 10:1 reduction of probe force yet Breakthrough results with 10:1 reduction of probe force yet achieving consistent and low contact resistanceachieving consistent and low contact resistance
1 – Prober Chuck Moves in Z-axis2 – Contact with Flip Chip Bump3 – Chuck moves in X-Y-Z
1
2
3
7SouthWest Test Conference June, 2003
MicroForce™ Probing on MicroForce™ Probing on PbPb//SnSn BumpBump
MicroSpring™ ContactProbing onHigh-Pb Bump
MicroForce™ Probing Motion
Chuck Translation Direction
FormFactor Laboratory DevelopmentFormFactor Laboratory Development
9SouthWest Test Conference June, 2003
TEL Synchronous XYZ DriveTEL Synchronous XYZ Drive
3D drive control method(Synchronous 3D Drive)(Synchronous 3D Drive)
TargetPosition
TsTime
1Drive
Start
Servo DriverServo Driver
Stepping Driver M
ControlPosition
Data
M
MHOST
XYZ Motion Contact
Benefits• Increased throughput with simultaneous XYZ drive• Smooth drive creates desired probe contact
SimultaneousXYZ Drive
10SouthWest Test Conference June, 2003
Experimental Setup Experimental Setup -- Predictive ProbingPredictive Probing
Force/Displacement Fixture
Test Wafer
Micro-ohmmeter(4-wire)
Data Acquisition
ForceDisplacement
Resistance
Test Spring
11SouthWest Test Conference June, 2003
Significantly Lowered Probe Force with Significantly Lowered Probe Force with MicroForce FeatureMicroForce Feature
0
2
4
6
8
10
12
14
0 0.001 0.002 0.003
Over-Travel (Inch)
Load
(Gra
ms)
S.R. = 0.0S.R. = 0.5S.R. = 1.0S.R. = 1.5S.R. = 2.0
Scrub Ratio =Movement in X/YMovement in Z
12SouthWest Test Conference June, 2003
Repeatability of Cres Study on Pb/Sn BumpsRepeatability of Cres Study on Pb/Sn Bumps
Stable and Low Contact Resistance with MicroForce Probing
Sort Floor ValidationSort Floor Validation
14SouthWest Test Conference June, 2003
Outline of Outline of PProbe card robe card CCharacteristics haracteristics MMeasurement easurement SSystemystem
DefinitionPCMS is a measurement system that can measure the probe card displacement under load.The result of PCMS includes the displacement of head plate, probe card and probe card holder.
0.0 um0.0 um
0.0 kg
Test Head
Probe card
Load sensor
Digital displacement sensor (Multiple)Chuck
Guide rail ofalignment bridge
Front sensor
Rear sensor
Leftsensor
Rightsensor
Sensor layout
Head plate
15SouthWest Test Conference June, 2003
Low Probing Force Using MicroForce FeatureLow Probing Force Using MicroForce Feature
• Used TEL’s PCMS to evaluate probe card deflection under varying loads. Data generated included actual spring compression and probing force per dialed chuck motion.
• Max probing force = 13N at 80µm of spring over-travel
• Max force/probe = 0.3g/mil with MicroForce vs. 3 – 4g/mil without MicroForce
0
2
4
6
8
10
12
14
0 10 20 30 40 50 60 70 80 90
O T ( u m )
Forc
e ( N
)
A vg F orc e
16SouthWest Test Conference June, 2003
Probing of Microprocessor DevicesProbing of Microprocessor Devices
TestingTesting–– Sort tests were run with and without MicroForce probing Sort tests were run with and without MicroForce probing
–– Correlation of device binning and other parametric indicators Correlation of device binning and other parametric indicators measuredmeasured
–– 3 production lots of microprocessor devices were tested3 production lots of microprocessor devices were tested
ResultsResults–– Binning data collected under MicroForce had a correlation rate oBinning data collected under MicroForce had a correlation rate of f
95.9 95.9 –– 97.4% vs. existing probing Recipe (target is >95%)97.4% vs. existing probing Recipe (target is >95%)
–– Other parametric measurements indicated comparable resultsOther parametric measurements indicated comparable results
–– No issues in terms of the mechanical and electrical behavior of No issues in terms of the mechanical and electrical behavior of the springs under MicroForce Probing. Metrology data remained the springs under MicroForce Probing. Metrology data remained within the defined specifications.within the defined specifications.
17SouthWest Test Conference June, 2003
Probe Mark ComparisonProbe Mark Comparison
Pb/Sn BumpPb/Sn Bump
Standard Probing MicroForce Probing
18SouthWest Test Conference June, 2003
Validation of MicroForce Low CresValidation of MicroForce Low Cres
0.08980.0898
0.12560.1256
0.14920.1492
0.34660.3466
Mean Mean
0.0200.020
0.0240.024
0.0230.023
0.0470.047
Std DevStd Dev
7070
5050
3535
1515
POT umPOT um• Good convergence of Cres
• Mean Cres at different programmable ovetravels: 0.1 – 0.35Ω
• Enabling a wide manufacturing process window
0
0.1
0.2
0.3
0.4
0.5
35µm POT0
0.1
0.2
0.3
0.4
0.5
50µm POT0
0.1
0.2
0.3
0.4
0.5
70µm POT0
0.1
0.2
0.3
0.4
0.5
Ω
15µm POT
19SouthWest Test Conference June, 2003
MicroSpring Contact After Aging TestMicroSpring Contact After Aging Test
After 500,000 Touchdowns
New (after probing 6 wafers)
20SouthWest Test Conference June, 2003
Cres Stability After Aging TestCres Stability After Aging TestUnder MicroForce Probing, Cres remained stable for 500K touchdowns
0.000
0.100
0.200
0.300
0.400
0.500
0.600
1 525 1049 1573 2097 2621
Die, Wafer
Die
Ave
Cre
s ( O
hm )
5k TD 100k TD 200k TD 300k TD 400k TD 500k TD
21SouthWest Test Conference June, 2003
ConclusionsConclusions
MicroForce Technology provided new capabilities to meet MicroForce Technology provided new capabilities to meet newnew--generation probing requirements. generation probing requirements.
MicroForce Probing offered low contact resistance necessary MicroForce Probing offered low contact resistance necessary for high power delivery requirements.for high power delivery requirements.
MicroForce Probing delivered very low probing force (<2g) MicroForce Probing delivered very low probing force (<2g) thus minimizing the risk to damage die with lowthus minimizing the risk to damage die with low--k ILD k ILD materials.materials.
Sort on production devices proved very stable and correlated Sort on production devices proved very stable and correlated to existing probing recipes. The solution satisfies Intel to existing probing recipes. The solution satisfies Intel probing requirements.probing requirements.
Future work will be targeted at:Future work will be targeted at:–– Throughput OptimizationThroughput Optimization–– Production Sort Certification Production Sort Certification
22SouthWest Test Conference June, 2003
AcknowledgmentAcknowledgment
The authors would like to thank the following contributors who The authors would like to thank the following contributors who made the MicroForce Probing development possible:made the MicroForce Probing development possible:
Tim CooperTim Cooper FormFactorFormFactor
GeneGene Kochert Kochert IntelIntel
Koichi Koichi Matsuzaki Matsuzaki TELTEL
Lance MilnerLance Milner IntelIntel
JesseJesse Nuanez Nuanez IntelIntel
Weida QianWeida Qian IntelIntel
Scott Scott ScofieldScofield IntelIntel
Rick Rick TakebuchiTakebuchi TELTEL
Dilip WadhwaniDilip Wadhwani IntelIntel
SunilSunil WijeyesekearaWijeyesekeara FormFactorFormFactor