Embed Size (px)
Citation preview
Dexter Electronic Materials / LoctitePage 1
No-Clean Assembly Process Conditions –Effects on Flip-Chip/Underfill Reliability
Michael Todd and Kathryn CostelloLoctite / Dexter Electronic Materials
IMAPS SoCal‘ 01 Conference
Dexter Electronic Materials / LoctitePage 2
Background
❚ Effects of Flux Residues on Underfill Properties❚ Tg Reduction❚ Flux induced voiding and flow striations❚ Reduction of interfacial adhesive strength
100105110115120125130135140145150
Tg, C
(TM
A)
No Flux I II III IV V
No Clean Flux
0
10
20
30
40
50
60%
Del
amin
atio
n (C
SAM
)
No Flux I II III IV V
No Clean Flux
96 Hrs Autoclave
Dexter Electronic Materials / LoctitePage 3
❚ Optimum package performance is dependent upon
underfill chemistry and flux chemistry –
❚ What processing parameters influence performance?
Underfill A Underfill B Underfill C
Background
Dexter Electronic Materials / LoctitePage 4
❚ Evaluate Effects of No-Clean Assembly Process Conditions on Flip-Chip Reliability
❚ Process Variations❚ Effect of No-Clean Flux Chemistry❚ Effect of Peak Assembly Reflow Temperature❚ Effect of Underfill Dispense Temperature
❚ Reliability Test Criteria❚ JEDEC Level 3 / 220°C Reflow
Experiment Goals
Dexter Electronic Materials / LoctitePage 5
Experimental Design
❚ 3 x 2 x 2 Full Factorial Design❚ Three Commercial No-Clean Fluxes❚ Two SMD Reflow Profiles
❚ Peak 220°C and Peak 260°C❚ Two Underfill Dispense Temperatures
❚ 90°C Preheat and 130°C Preheat
❚ Four Test Assemblies per Condition (Screening Evaluation Only)
Dexter Electronic Materials / LoctitePage 6
Material Sets Evaluated
FluxFormulation
Chemistry Viscosity(cps)
Residue
A Synthetic 110,000 4%
B NotReported
10,000 Medium
C Synthetic 30,000 Very Low
Underfill Epoxy-Anhydride
5,000
Dexter Electronic Materials / LoctitePage 7
Test Vehicle
❚ 10mm x 10mm Die❚ Silicon Nitride Passivation❚ Perimeter and Internal I/O
❚ 0.79mm Thick FR4 Laminate
❚ Cu/Ni/Au Plating❚ 4 Die / Substrate
Dexter Electronic Materials / LoctitePage 8
Assembly Process
❚ Prebake Substrate at 125°C❚ Die dip-fluxed at 0.0015” thickness❚ Die placed onto substrate❚ Die reflow soldered using 220°C or 260°C
SMD Profile❚ (No cleaning operation)❚ Underfill dispensed at 90°C or 130°C❚ Underfill cured 30 Minutes at 165°C
Dexter Electronic Materials / LoctitePage 9
Evaluation Process
❚ CSAM Immediately Following Underfill Process
❚ Expose Parts to JEDEC L3 Preconditioning❚ 30°C/85% RH/192 Hours❚ 3X Reflow Cycles (Peak Temp = 220°C)
❚ CSAM
Dexter Electronic Materials / LoctitePage 10
Results - Initial ObservationsEffect of Flux Composition
Flux A Flux B Flux C220°C Reflow Process
90°C Underfill Dispense Temperature
Flux Residue “Shadow”
Dexter Electronic Materials / LoctitePage 11
Results - Initial ObservationsEffect of Reflow Temperature
Flux B220°C Reflow Process90°C Underfill Dispense
Temperature
Flux B260°C Reflow Process90°C Underfill Dispense
Temperature
Flux Residue “Shadow” Flow Void
Dexter Electronic Materials / LoctitePage 12
Results - Initial ObservationsEffect of Underfill Dispense Temperature
Flux C260°C Reflow
Process130°C Underfill
Dispense
Flux C260°C Reflow
Process90°C Underfill
Dispense
Flux Residue “Shadow”
Dexter Electronic Materials / LoctitePage 13
Results - Post JEDEC L3220°C Assembly Reflow Temperature
Flux A220°C Reflow
Process90°C Underfill
Dispense
Flux B220°C Reflow
Process90°C Underfill
Dispense
Flux C220°C Reflow
Process90°C Underfill
Dispense
Dexter Electronic Materials / LoctitePage 14
Results - Post JEDEC L3260°C Assembly Reflow Temperature
Flux A260°C Reflow
Process90°C Underfill
Dispense
Flux B260°C Reflow
Process90°C Underfill
Dispense
Flux C260°C Reflow
Process90°C Underfill
Dispense
Dexter Electronic Materials / LoctitePage 15
Experiment Conclusions
❚ Effects of Flux Residue❚ Shadow Images of Flux Residue Observed
by CSAM Analysis❚ Effects of Reflow Process
❚ Flow Voids Caused by Higher Residue Flux after 260°C Reflow Process
❚ All Samples Passed JEDEC L3 after 220°C Assembly Reflow Process
❚ Only One Flux Sample Passed JEDEC L3 after 260°C Assembly Reflow Process
Dexter Electronic Materials / LoctitePage 16
Experiment Conclusions
❚ Effects of Dispense Temperature❚ Higher Dispense Temperature Reduced Flux
Shadowing Only on One Flux Chemistry❚ No Impact on Most Flux Chemistries
Dexter Electronic Materials / LoctitePage 17
Follow-Up Evaluations
❚ FTIR Evaluation of the Effects of Reflow Temperature on Flux Residue
Dexter Electronic Materials / LoctitePage 18
Evaluation of “Shadow” Regions
0
50
100
150
200
250
-0.06
-0.05
2-0.
044
-0.03
6-0.
028
-0.02
-0.01
2-0.
004
0.004
0.012 0.0
20.0
280.0
360.0
440.0
52 0.06
0.068
0.076
0.084
0.092 0.1
0.108
0.116
0.124
0.132 0.1
40.1
480.1
560.1
640.1
72 0.18
0.188
0.196
0.204
0.212 0.2
20.2
280.2
360.2
440.2
52
Time of Flight (µsec)
Am
plitu
de
0
50
100
150
200
250
-0.06
-0.05
2-0.
044
-0.03
6-0.
028
-0.02
-0.01
2-0.
004
0.004
0.012 0.0
20.0
280.0
360.0
440.0
52 0.06
0.068
0.076
0.084
0.092 0.1
0.108
0.116
0.124
0.132 0.1
40.1
480.1
560.1
640.1
72 0.18
0.188
0.196
0.204
0.212 0.2
20.2
280.2
360.2
44
Time of Flight (µsec)
Am
plitu
de
0
50
100
150
200
250
-0.06
-0.05
2-0.
044
-0.03
6-0.
028
-0.02
-0.01
2-0.
004
0.004
0.012 0.0
20.0
280.0
360.0
440.0
52 0.06
0.068
0.076
0.084
0.092 0.1
0.108
0.116
0.124
0.132 0.1
40.1
480.1
560.1
640.1
72 0.18
0.188
0.196
0.204
0.212 0.2
20.2
280.2
360.2
440.2
52
Time of Flight (µsec)A
mpl
itude
Dexter Electronic Materials / LoctitePage 19
Evaluation of “Shadow” Regions
0
50
100
150
200
250
-0.06
-0.05
2-0.
044
-0.03
6-0.
028
-0.02
-0.01
2-0.
004
0.004
0.012 0.0
20.0
280.0
360.0
440.0
52 0.06
0.068
0.076
0.084
0.092 0.1
0.108
0.116
0.124
0.132 0.1
40.1
480.1
560.1
640.1
72 0.18
0.188
0.196
0.204
0.212 0.2
20.2
280.2
360.2
440.2
52
Time of Flight (µsec)
Am
plitu
de
Delamination Flux / UnderfillInterface
70µ
Dexter Electronic Materials / LoctitePage 20
Evaluation of “Shadow” Regions
Underfill Level GateFlux Level Gate
