2017 nepp tasks update for ceramic and …...2017 nepp tasks update for ceramic and tantalum...

2017 NEPP Tasks Update for Ceramic and Tantalum 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

To be presented by A.Teverovsky at the NASA Electronic Parts and Packaging (NEPP) Electronics Technology Workshop, Greenbelt, MD, June 2017.

List of Acronyms


AF acceleration factor MLCC multilayer ceramic capacitorBME base metal electrode MOR modulus of ruptureDCL direct current leakage PME precious metal electrodeESR Equivalent series resistance QA quality assurance

FPGA field-programmable gate array RB reverse biasHALT highly accelerated life testing S&Q screening and qualification

HT High temperature SMT surface mount technologyHTS high temperature storage TC temperature cyclingIDC inter-digitated capacitor VH Vickers hardnessIFT Indentation Fracture Test WTC wet tantalum capacitor

2

Outline Update on tantalum capacitors. Leakage currents, gas generation and case

deformation in wet tantalum capacitors. MnO2 chip capacitors:

• ESR degradation.• Acceleration factors for DCL degradation and failures.• Effect of moisture on degradation of reverse currents. Polymer capacitors. Future work.

Update on ceramic capacitors. Mechanical properties of MLCCs. Failures in BME capacitors with defects. Effect of cracking on degradation of MLCCs at HT. Future work.


3

Leakage Currents, Gas Generation and Case Deformation in Wets


4

NEPP report (https://nepp.nasa.gov/) contains: Part I. Analysis of leakage currents; Part II. Gas generation, hermeticity, and pressure in the case. Part III. Electrolyte at the glass seal. Part IV. Deformation of cases in high capacitance value wet tantalum capacitors.

Risks of internal leaks: non-oxidized surfaces; corrosion of welds; excessive leakage currents and gas generation.

To reduce failures a special conditioning at HT is recommended.

WTCs with different types of internal seals

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E+2 1.E+3 1.E+4 1.E+5 1.E+6

curr

ent,

A

time, sec

DWG04005 240uF 125V

Anomalies in leakage currents indicate presence of defects

https://nepp.nasa.gov/

Leakage Currents, Gas Generation and Case Deformation in Wets, Cont’d

To be presented by Alexander Teverovsky at the 2017 NASA Electronics Parts and Packaging (NEPP) Electronics Technology Workshop (ETW), NASA Goddard Space Flight Center, Greenbelt, MD, June 26-29, 2017.

5

Multiple (1650) transients did not cause case bulging likely due to H2 outdiffusion through tantalum case.

TC can result in irreversible lid deformation and excessive DCL. To assure reliable operation in vacuum, HTS testing at 150 ºC for

1000 hours is recommended. Glass seal protection in button case capacitors is less effective

compared to the cylinder case parts.

y = 4.2153e0.036x

0

500

1000

1500

2000

2500

3000

0 50 100 150 200

disp

lace

men

t, um

temperature, deg.C

HE3 SN5

0

20

40

60

80

100

120

140

160

0

100

200

300

400

500

600

700

800

0 100 200 300 400 500 600

tem

pera

ture

, deg

.C

disp

lace

men

t, um

time, min

HE3 SN5

displ, umtemp, C

Cracking of the seal after exposure to HT

Bulging of the case

ESR Degradation in MnO2 CapacitorsA report (https://nepp.nasa.gov/) includes analysis of environmental factors: vacuum, high temperature storage, temperature cycling, moisture, and soldering.


6

0

50

100

150

200

250

300

0 200 400 600 800 1000

ESR

, mO

hm

time, hr

HTS 150C CWR29FC686KBGA, DC0703, ESR_max=275 mohm

0

20

40

60

80

100

120

0 200 400 600 800 1000

ESR

, moh

m

time, hr

220uF 10V Mfr.B at 85C 85%RH

Most parametric ESR failures are due to insufficient margin to ESRlimit.MnO2 caps can withstand 1000 hr at 150 °C and at 85 °C/85% RH.AEC-Q200 requirements are much more severe compared to M55365.Compressive stresses after bake reduce delaminations and squeeze

microcracks in cathode layers resulting in reduction of ESR.Swelling of MC and stress relaxation in moisture have opposite effects.

Examples of ESR variations during HTS and humidity testing


Acceleration Factors for DCL Degradation and Failures


7

A report (https://nepp.nasa.gov/) describes catastrophic and parametric failures in Ta capacitors, their mechanisms and AF.

Analysis showed that 5.5 < B < 10.3, 1.42 < Ea <1.66 eV. Parametric degradation is reversible and can be annealed at HT. The mechanism of degradation is attributed to migration of

oxygen vacancies in the dielectric with E ~1.1 eV.

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

0 500 1000 1500 2000

curr

ent,

A

time, hr

6.8uF 25V Mfr.A at 125C, 16.6V

SN1 SN2SN3 SN4SN5 SN6Failure

Simulation of TTFs for 6.8uF 25V capacitors

time, hr

cumu

lative

prob

abilit

y, %

1.E+2 1.E+91.E+3 1.E+4 1.E+5 1.E+6 1.E+7 1.E+81

5

10

50

90

99

experimental data

calculation

125C, 16.6V

use conditions55C, 10V

1.E-09

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

init HALT125C35V

bake20hr175C

HALT145C25V

bake10hr175C

bake30hr175C

bake60hr175C

curr

ent@

25V

, 85C

, A

6.8uF 25V capacitors

−×= 1exp

VRVBAF test

V

−×−=

21

11expTTk

EAT aT


Kinetics of Moisture Sorption in MnO2 Capacitors

8To be presented by Alexander Teverovsky at the 2017 NASA Electronics Parts and Packaging (NEPP) Electronics Technology Workshop (ETW), NASA Goddard Space Flight Center, Greenbelt, MD, June 26-29, 2017.

( ) ( )

−−−×−+= τ

DttCCCtC exp1)( minmaxmin DdtD ×

=6

2

Spd××

= ~γτ

-0.5

0.5

1.5

2.5

3.5

4.5

0.1 1 10 100 1000

∆C/C

min

, %

time, hr

Capacitors at 22C 85%RH after bake

CC1calcCC6calcMC1calc

10

10.1

10.2

10.3

10.4

0.1 1 10 100

capa

cita

nce,

uF

time, hr

10uF MLCCs at 125C

CC1

MC1

E = 0.5 eV

Ea =0.43eV

0.1

1

10

100

0.002 0.0025 0.003 0.0035 0.004

diffu

sion

del

ay, h

r

1/T, 1/K

CC1

CC6

MC1

MC

CC

Ea=0.36 eV

Ea= 0.33 eV

1

10

100

1000

0.002 0.0025 0.003 0.0035 0.004

char

acte

ristic

tim

e, h

r

1/T, 1/K

CC1CC6MC1

MC

CC

Slugs in tantalum chip capacitors can be used as moisture sensors. A model for C-tvariations has been developed. Bake-out times can be selected based on the characteristic times of the desorption process.

Moisture sorption can be characterized by two time constants

Effect of Moisture on Degradation of Reverse Currents in MnO2 Capacitors


0

5

10

15

20

25

30

35

0 200 400 600 800 1000 1200 1400 1600

curre

nt, m

A

time, hr

194D at RBS 5V, 75C

air

vacuum

194D effect of humidity at 22C and 4V RB

steady-state current, mA

cumu

lative

prob

abilit

y, %

1.E-2 101.E-1 11

5

10

50

90

99

7% RH

35% RH

85% RH

Degradation under RB strongly depends on presence of moisture in environments and preconditioning.

Oxygen vacancies play important role in formation of protonic species.

xOiO OHVOH +→+ ••• 22••• →++ O

xOO OHOVOH 22

RBS of 10uF 25V capacitors

Polymer Capacitors


Report “Evaluation of 10V chip polymer tantalum capacitors…” (https://nepp.nasa.gov/) describes issues with low-voltage parts.

6.0E-8 A/hr

1.4E-07 A/hr

1.0E-07 A/hr

0.0E+0

5.0E-6

1.0E-5

1.5E-5

2.0E-5

2.5E-5

3.0E-5

0 20 40 60 80 100 120 140 160

curr

ent,

A

time, hr

T530 330uF at 125C 10V

SN2

SN7

SN9 1

10

100

1000

0 2500 5000 7500 10000

ESR

, moh

m

time, hr

HTS at 100C

T510 330uF

T530 150uF

T525 220uF

0.0E+0

5.0E-5

1.0E-4

1.5E-4

2.0E-4

2.5E-4

3.0E-4

3.5E-4

0 200 400 600 800 1000

curre

nt, A

time, hr

220uF 10V at 85C, 10V

Parametric failures are likely to be major issues with polymer caps.QA system developed for MnO2 capacitors is not applicable.Unstable leakage currents that might be more significant in vacuum.Significant degradation of ESR during HTS.


Polymer Capacitors, Cont’d


One of the most intriguing problems: anomalous transients after HTS

Moisture plays an important role in the mechanism of transients.

Anomalies in long-term transient currents might be due to changes in the trap system of Ta2O5.

More analysis necessary.

1.E-7

1.E-6

1.E-5

1.E-4

1.E-3

1.E-2

1.E-1

-80 -60 -40 -20 0 20 40 60 80 100 120 140

curr

ent,

A

temperature, deg.C

22uF 25V virgin and after HTS at 25V 100sec

virgin

HTS

1.E-8

1.E-7

1.E-6

1.E-5

1.E-4

1.E-3

1.E-2

1.E-1

-80 -60 -40 -20 0 20 40 60 80 100 120 140

curr

ent,

A

temperature, deg.C

22uF 25V virgin and after 85C/85%RH at 25V 100sec

HUM

virgin

1.E-7

1.E-6

1.E-5

1.E-4

1.E-3

1 10 100 1000

curr

ent,

A

time, sec

22uF 25V HTS at 125C

SNu init 70SNu 23hr70SNu 95hr70SNu 311hr70SNu 1007hr70SNu 3009hr70SNu 4012hr70SNu 4996hr70

0

10

20

30

40

50

60

0 10 20 30

Vol

tage

, V

time, sec

22uF 25V capacitors after 120hr HTS 125C or 85C/85%RH HTS 1mA

HTS 2mA

HTS 3mA

HTS 4mA

HTS 5mA

HTS 6mA

HTS 8mA

HTS 10mA

HTS 12mA

HTS 14mA

HUM 0.1mA

Future Work on Tantalum Capacitors


MnO2 chip capacitors. Complete current tasks. Reliability acceleration factors for automotive grade capacitors.

Advanced wet capacitors. Effect of HT storage on performance and reliability. Evaluation of SMT wet tantalum capacitors.

Polymer capacitors. Degradation models for HTS and recommendations for S&Q.

Super-capacitors for space application.

12

Mechanical Properties of MLCCs


Can mechanical characteristics predict robustness of MLCCs under soldering stresses?

Report that is available at https://nepp.nasa.gov/ includes: Flexural Strength Testing of MLCCs. Vickers Hardness Testing. Indentation Fracture Test (IFT).

223bdFLMOR =

AEC-Q200-003

CDR35BX104AKUS, 0.1uF, 50V, size 1825, Mfr.C

MOR, MPa

cumu

lative

prob

abilit

y, %

40 4001001

5

10

50

90

99

initialactivated rosin fluxpolishedmoisture: 160hr, 85% RH, 22Cmoisture: 160hr, 85% RH, 22CTSD350 & moisture

10N5N 2.5N

47

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

Flexural strength method determines tensile strength at the surface. No substantial difference between mechanical characteristics of BME and PME

capacitors. Smaller size MLCCs have greater strength – Benefits of BMEs. Same size capacitors can be used for comparative analysis of the lots. Variations of MOR values from lot to lot might exceed 50%.

223bdFLMOR =Modulus of rupture


Vickers Hardness


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 significant difference between PME and BME capacitors. Improvements to reduce errors might allow for revealing

differences in lots.

Hardness is a resistance to indentation.

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


IFT can provide useful information regarding robustness of capacitors under soldering conditions, but additional analysis is necessary.

Mechanical testing might be useful for selecting robust parts for manual soldering, but more work is necessary to reduce errors and select criteria.

For critical applications a combination of assembly simulation and special testing might be recommended.

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

0

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

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

PN 1 PN 2

Kc, M

pa_m

^0.5

PME_V capacitors

SN1 SN2 SN3 SN4 SN5

Failures in BME Capacitors with Defects


Migration of VO++ is enhanced either by increased E, as in case of

thinning of the dielectric, or by increased µ, as in case of cracks. Catastrophic failures occur when ΦB decreases to ΦBcr. For

capacitors having small, micrometer-size defects ΦBcr is low so catastrophic failures are unlikely.

In the range of typical HALT conditions voltage increases the probability of catastrophic failures to a greater degree compared to temperature. This might result in errors in AFs.

Thermal run-away model

),,(2Bddd ETJrI Φ××= πθRTTVI

dtdTC d

0−−×=×

Φ−=

kTE

kTEATJ sB

S

5.02/3 expexp βµ


Stretched exponential dependence of ΦB with time.In bulk: τV = 200 hr, (µV ~5×10-15 cm2/Vs)Along the crack: τc ~10 hr (µV ~10-13 cm2/Vs).

Model: accelerated migration of VO++ along the cracks.

Simulations are in reasonable agreement with experimental data.

−−×∆Φ=Φ

γ

τVBB

tt exp1)(

Vd

VV ×

×=

µτ

278.0

0

0.15

0.3

0.45

0.6

0.75

0.9

1.E-8

1.E-7

1.E-6

1.E-5

1.E-4

1.E-3

1.E-2

0.1 1 10 100 1000

∆ΦB,

eV

curre

nt, A

time, hrI defect free I defect 1I total def. 1 I defect 2I total def. 2 DFB_def 1 =0.45eVDFB_def 2 =0.55eV DFB =0.1eV

∆ΦBcr =0.54eV

Effect of Cracking on Degradation of MLCCs at HT

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

0.1 1 10 100 1000

curre

nt, A

time, hr

virginfractured

0.33uF/50V at life test

cracks

virgin

Simulation of life testing

Future Work on Ceramic Capacitors

Evaluation of IDC capacitors used for FPGAs.

Comparative analysis of performance and reliability of BME and PME capacitors. Breakdown voltages. Analysis of failures in BME capacitors with defects. Express testing to determine reliability acceleration factors for BME

capacitors. Guidelines for selecting “auto” MLCCs for different project levels.


18

2017 nepp tasks update for ceramic and …...2017 nepp tasks update for ceramic and tantalum...

Documents