Cracking Problems and Mechanical Characteristics of PME and BME
Ceramic Capacitors
Alexander TeverovskyAS&D, Inc.
Work performed for Parts, Packaging, and Assembly Technologies Office,
NASA GSFC, Code [email protected]
NASA Electronic Parts and Packaging (NEPP) Program
Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
List of Acronyms
Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
2
BME base metal electrode IFT Indentation Fracture Test
C-SAM C-mode scanning acoustic microscopy IR insulation resistance
DCL direct current leakage IWT ice water test
DF dissipation factor MLCC multilayer ceramic capacitor
ECM electrochemical migration MOR modulus of rupture
EDS energy dispersive spectroscopy PME precious metal electrode
EM electrical measurements RH relative humidity
ESR Equivalent series resistance TSD terminal solder dip
FA failure analysis VH Vickers hardness
Abstract
Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
3
Most failures in MLCCs are caused by cracking that create shorts betweenopposite electrodes of the parts. A use of manual soldering makes this problemespecially serious for space industry. Experience shows that different lots ofceramic capacitors might have different susceptibility to cracking under manualsoldering conditions. This simulates a search of techniques that would allowrevealing capacitors that are most robust to soldering-induced stresses.Currently, base metal electrode (BME) capacitors are introduced to high-reliabilityapplications as a replacement of precious metal electrode (PME) parts.Understanding the difference in the susceptibility to cracking between PME andBME capacitors would facilitate this process.This presentation gives a review of mechanical characteristics measured in-situon MLCCs that includes flexural strength, Vickers hardness, indentation fracturetoughness, and the board flex testing and compare characteristics of BME andPME capacitors. A history case related to cracking in PME capacitors thatcaused flight system malfunctions and mechanisms of failure are considered.Possible qualification tests that would allow evaluation of the resistance ofMLCCs to manual soldering are suggested and perspectives related tointroduction of BME capacitors discussed.
Outline
Problems with cracking of MLCCs. In-situ mechanical testing.
Flexural strength. Vickers hardness. Indentation fracture toughness. Board flex testing.
Failure history case. What can be done to mitigate manual soldering
cracking. Conclusion.
Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
4
Cracking-Related Problems Mechanism of cracking during manual soldering. Revealing cracks in capacitors (loose and soldered).
Electrical measurements (C, DF, IR, VBR) Electromechanical effects Visual, radiography, ultrasonic analysis
Effects of cracking on reliability in humid and HT environments. Robustness of MLCCs towards thermo-mechanical stresses.
IWT (ice water testing) TSD (terminal solder dip testing)
In-situ mechanical testing. Flexural strength Vickers hardness Indentation Fracture Test Board flex testing
Susceptibility to cracking of PME and BME capacitors.
Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
5
Do different types of MLCCs have different susceptibility to cracking under manual soldering conditions and how it can be revealed?
Degradation of MLCCs with cracks and the effectiveness of different techniques is described in various publications and reports posted at the NEPP web site.
Flexural Strength
6Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
The test is described in AEC-Q200, but acceptance criteria are up to the users.
223bdFLMOR =
AEC-Q200-003
CDR35BX104AKUS, 0.1uF, 50V, size 1825, Mfr.C
MOR, MPa
cumu
lative
pro
babil
ity, %
40 4001001
5
10
50
90
99
initialactivated rosin fluxpolishedmoisture: 160hr, 85% RH, 22Cmoisture: 160hr, 85% RH, 22CTSD350 & moisture
10N5N 2.5N
No effect of possible flaws but surface cracks reduce MOR. Smaller size MLCCs have greater strength – Benefits of BMEs. No substantial difference between BME and PME capacitors. Variations of MOR values from lot to lot might exceed 50%. The test can be used for relatively large (≥1206) parts. Same size capacitors can be used for comparative analysis.
Modulus of rupture 4
7
4 7 19 13
58
8
0
100
200
300
400
500
600
1206 1210 1808 1812 1825 2225
char
acte
ristic
MO
R, M
Pa
case size
Effect of case size on MOR
PME
BME
3
Vickers Hardness
7Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
In-situ VH measurements are possible using MLCCs with relatively thick cover plates. P should be low so the depth of the indentation is < 2x the thickness of the cover plate.
No correlation between strength and VH for ceramic materials. No significant difference between PME and BME capacitors. Reduction of errors might allow for revealing differences in lots.
Hardness is a resistance to indentation.Testing is specified in ASTM C1327-15 (2015)
2854.1
DPVH ×
=
Vickers Hardness Test for PME and BME Capacitors
VH, GPa
cumu
lative
pro
babil
ity, %
7 128 9 10 111
5
10
50
99
BME
PME
90% confidence
y = 191.65x
0
100
200
300
400
500
600
700
0 1 2 3 4
D^2
, um
^2
load, N
Vickers test, LT capacitors
SN1
SN2
SN3
SN4
SN5
Indentation Fracture Test
8Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
Test results depend on environments and time of exposure. IFT might be useful for selecting parts for manual soldering, but
more work is necessary to reduce errors and select criteria. No significant difference between PME and BME capacitors.
Fracture Toughness: the ability of a material to withstand stresses in the presence of cracks.
IFT technique is the most controversial.
= 5.1
5.0
_ cP
VHEIFT MRξ
PME BX
PME BP
BME X7R
IFTavr 1.08 1.46 0.95STD 0.20 0.19 0.23
N 12 2 60
20
40
60
80
100
120
140
160
180
0 1 2 3 4 5
c^1.
5, (u
m)^
1.5
load, N
PME_L capacitors
PME_L1, 1.7
PME-L2, 1.32
PME_L3, 1.49
PME_L4, 1.20
1.03 1.07
0.83
1.35
1.000.89
1.33
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
Kc,
Mpa
_m^0
.5
Different lots of capacitors
L1
L2
L3
Board Flex Testing
9Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
AEC-Q200-005 specifies conditions and the size of the test board. “A failure is when a part cracks or causes a change in the parametric being monitored.”
M32535 allows for multi-chip boards. Failure criteria: C >+/-10% at δ = 2 mm.
Factors affecting test results: Orientation of the component; Attachment with Ag-epoxy absorbs stress; Solder fillet height, and thickness under the chip; Solder type (less cracking for Pb-free alloys) MLCC material (X7R weaker than COG) Larger chips experience greater stress and have
greater susceptibility to cracking. This test is widely used to address cracking during de-panelization. Results are affected by variety of factors. Conditions for using multi-chip boards need additional analysis.
Board Flex Testing, Cont’d
10Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
Mechanical stresses affect characteristics of class II capacitors. Deviations of ε can change C during flex testing up to a few %. Variations of C might be reversible even in the presence of cracks.
0
1
2
3
4
5
6
00.05
0.10.15
0.20.25
0.30.35
0.40.45
0.5
0 50 100 150 200
defo
rmat
ion,
mm
C, u
F
time, sec
PME 2225 X7R 0.47uF 50V A2 SN1
C, uF
bend, mm
0
1
2
3
4
5
6
9.40E-2
9.45E-2
9.50E-2
9.55E-2
9.60E-2
9.65E-2
0 50 100 150 200
defo
rmat
ion,
mm
C, u
F
time, sec
BME 0603 X7R 0.1uF 50V SN2
C, uF
bent, mm
0
0.5
1
1.5
2
2.5
3
3.5
10.4
10.5
10.6
10.7
10.8
10.9
11
11.1
0 50 100 150
defo
rmat
ion,
mm
C, n
F
time, sec
PME 0805 X7R 0.01uF 25V SN1
C, uFbend, mm
0
0.5
1
1.5
2
2.5
3
3.5
2.22.222.242.262.282.3
2.322.342.362.382.4
0 50 100 150
defo
rmat
ion,
mm
DF,
%
time, sec
PME 0805 X7R 0.01uF 25V SN1
DF, %
bend, mm
0
0.5
1
1.5
2
2.5
3
3.5
-1.0E-09
-5.0E-10
0.0E+00
5.0E-10
1.0E-09
1.5E-09
2.0E-09
2.5E-09
3.0E-09
0 100 200 300
defo
rmat
ion,
mm
curr
ent,
A
time, sec
PME 0805 X7R 0.01uF 25V SN1
I, A
bend, mm
Board Flex Testing, Cont’d
11Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
Variations of capacitance having cracks with load are reversible.Degradation of C by steps indicates partial separation of electrodes.Distributions of δcr (0.9Cinit) might be more effective for assessment
of the vulnerability to cracking compared to 2 mm pass-fail criteria.No substantial difference in the flex cracking between PME and
BME capacitors.A smaller size of BME compared to similar value PME capacitors
makes BME less vulnerable to flex cracking.
0
1
2
3
4
5
6
0
0.002
0.004
0.006
0.008
0.01
0.012
0 60 120 180 240de
form
atio
n, m
m
C, u
F
time, sec
PME 0603 X7R 0.01uF 25V SN2
C, uF
bend, mm
0
1
2
3
4
5
6
7
0.05
0.06
0.07
0.08
0.09
0.1
0.11
0 50 100 150 200 250
defo
rmat
ion,
mm
C, u
F
time, sec
BME 0603 X7R 0.1uF 50V SN1
C, uF
bend, mm
A History Case
12Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
On-orbit anomalies after months of operation were attributed to excessive leakage currents in CDR35 capacitors.
The parts were soldered manually and suspected of having cracks. Testing of a spare unit on the ground also showed increasing
leakage currents after several weeks of operation. FA: the failure was due to delaminations and cracking in the part. No external cracks on the failed lot were observed. Acoustic microscopy showed that a substantial proportion of parts
had delaminations at the termination areas.
Courtesy of L.Panashchenko and R.Weachock
A History Case. Test Plan.
13Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
To check whether the flight lot has high susceptibility to cracking and to identify techniques for assessment of the robustness of parts to manual soldering, 20 capacitors from the flight lot (lot A) and a reference lot (lot C) were tested in parallel.
o Before stress testing the parts were characterized by mechanical, electrical and acoustic (C-SAM) tests.o Terminal solder dip testing (TSD350) was used to simulate manual soldering stresses.o Leakage currents were monitored with time at different voltages and environmental conditions.
A History Case. Initial Characteristics.
14Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
Some difference between lots was revealed by acoustic microscopy and flexural strength testing.
CDR35 0.47uF 50V capacitors
MOR, MPa
cumu
lative
pro
babil
ity, %
50 2001001
5
10
50
90
99
Mfr.A
Mfr.C
(113/8.6 MPa)
(148/17.7 MPa)
Optical examination: no anomalies.EDS: similar composition of ceramics.C_SAM: no defects in the lot C and
corner delaminations in lot A.No substantial difference in distributions
of C, DF and IR between the lots.Both lots had DCL at 2VR <10-10 A.VH and IFT did not reveal significant
difference. However, the flexural strength was greater for lot C.
y = 2E-07x-1.071
y = 1E-07x-1.041
1.E-10
1.E-9
1.E-8
1.E-7
1.E-6
1.E+0 1.E+1 1.E+2 1.E+3
curr
ent,
A
time, sec
CWR35 0.47uF 50V Mfr.A
50V
0_50V
100V
0_100V
courtesy of C.Greenwell
A History Case. Effect of TSD350.
15Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
20 samples were subjected to 3 cycles of TSD at 350 ºC
Two lots had a substantially different propensity to cracking and electrical failures after manual soldering simulations.
visual C-SAM EM 1000hr at 50V, 22ºC/85%RH
Lot A 20/20 18/20 1/20 6/20Lot C 1/20 0/20 0/20 0/20
0.E+0
2.E-9
4.E-9
6.E-9
8.E-9
1.E-8
0 200 400 600 800 1000
curr
ent,
A
time, hr
Mfr.C after TSD350 and CSAM at 22C 85% RH, 50V
1.E-10
1.E-9
1.E-8
1.E-7
1.E-6
1.E-5
1.E-4
0 200 400 600 800 1000
curr
ent,
A
time, hr
Mfr.A after TSD350 and CSAM at 22C 85% RH, 50V
SN A1 SN A5SN A6 SN A9SN A15 SN A18SN A20 1 G_uF
Lot CLot ALot A
A History Case. Failure Analysis.
16Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
Cracks in lot A started from the surface and continued as metal/ceramic delamination.
To evaluate interaction of cracks with delaminations, virgin and post-TSD samples were fractured in the middle.
No delaminations on virgin samples from lot A and on both virgin and post-TSD samples from lot C.
TSD testing and fracturing resulted in delaminations located mostly at electrodes close to the surface of capacitors.
Fractured post-TSD samples
A History Case. Failure Mechanism.
17Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
Cracking and delaminations caused by electroplating occur more often with PME than with BME capacitors.
Probability of failures after manual soldering was increased due to the presence of delaminations.
Corner location indicates that delaminations might be due to formation of terminals.
Reasons for corner delaminations [*]: Thick Ni layers increase mechanical stresses.Generation of H2 during electroplating:• Decreases fracture toughness of ceramics;• Removes PdO barrier on Ag/Pd electrodes,
weakens the interface and facilitates ECM of Ag.• Fast evolution of H2 might cause pop-corning
during soldering.
electroplating
formed capacitor
soldering
[*] "Susceptibility to Cracking of Different Lots of CDR35 Capacitors,", NEPP report2017, https://nepp.nasa.gov/files/29210/NEPP-TR-2018-Teverovsky-CDR35-Capacitors-TN52049.pdf
Resistance to Manual Soldering Test
18Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
A combination of TSD350, C-SAM, EM, and humidity tests can be used to select parts more robust to manual soldering.
Depending on criticality some tests can be skipped or replaced.
Initial EM (C, DF, IR)• No failures allowed.• In case of failures →
reject or rescreen.
TSD350, 2 sides 3 cycles• If > 25% samples
have visual cracks → reject.
EM (C, DF, IR)• No failures allowed.• Analyze distributions
for outliers.
C-SAM• If a sample has >25%
delaminated area→ reject.• if > 25% samples have
corner cracks → reject.
Monitoring I-t at 85%RH, VR for 1000hr• If current spikes
exceed Icr → reject.
None of the tests provides reliable information regarding the susceptibility to cracking, but some test have better sensitivity.
Example of test sequence
Conclusions
19Presented by Alexander Teverovsky at the 2018 CMSE Components for Military & Space Electronics Training & Exhibition, Los Angeles, CA, May 7-10, 2018.
In-situ mechanical testing of MLCCs, and the flexural strength in particular, can reveal lot-to-lot variations.
Different lots of MLCCs do have different susceptibility to cracking and failures caused by manual soldering.
No substantial difference between mechanical characteristics of PME and BME capacitors.
Capacitors with cracks can pass ground phase testing, but fail during the mission.
Combination of TSD350, C-SAM and electrical testing can be used to mitigate the risk of failures.
Due to a smaller size and different degradation mechanisms, BME capacitors have a lesser probability of failures caused by manual soldering compared to PME parts. However, more problems might be expected with small (≤0603) size MLCCs.
Top Related