36
Ionic Migration on Printed-Circuit Boards ESPEC CORP. CONTENS 1. Introduction 2. About ionic migration 3. Test methods 4. Insulation resistance test (Copper and silver migration) 5. Solder alloy migration

Ionic Migration on Printed-Circuit Boards

  • Upload
    others

  • View
    7

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Ionic Migration on Printed-Circuit Boards

Ionic Migration on Printed-Circuit Boards

ESPEC CORP.

CONTENS

1. Introduction

2. About ionic migration

3. Test methods

4. Insulation resistance test

(Copper and silver migration)

5. Solder alloy migration

Page 2: Ionic Migration on Printed-Circuit Boards

Migration occurring on electronics components

Copper migration

Copper migration

(a) Copper migration on PCB (b) Copper migration on telephone connector

Gold plated copper terminal

Page 3: Ionic Migration on Printed-Circuit Boards

-20

0

20

40

60

80

100

2000/5/4

2000/5/14

2000/5/24

2000/6/3

2000/6/13

2000/6/23

2000/7/3

2000/7/13

2000/7/23

2000/8/2

Temp ( ℃) / Humi.(%RH)

-20

0

20

40

60

80

100

1999/12/21

1999/12/31

2000/1/10

2000/1/20

2000/1/30

2000/2/9

2000/2/19

2000/2/29

2000/3/10

Temp.(℃) / Humi (%RH)

Solder crack

Dew condensation

5mm

Connected deterioration

(a) Temperature and humidity in automobile (b) Failure on PCBs

winter

summer

Humi.

Temp.

Failure on PCB and environmental stress

Insulation deterioration

Migration occurring

Temp. cyclic

Page 4: Ionic Migration on Printed-Circuit Boards

Factors of failure on PCB

Failure phenomena

Connected deterioration

Insulation deterioration

Fatigue – solder clack

Oxidation – contact defective

Corrosion – migration

Adhesion – water absorption

Failure mode Factor phenomena result

Increase of contact resistance

Decrease in insulation resistance

Transform – decrease in insulation distance

Abrasion

Breakdown

Breakdown

Page 5: Ionic Migration on Printed-Circuit Boards

About ionic migration

Page 6: Ionic Migration on Printed-Circuit Boards

Classification of migration phenomenon

(a) Ionic migration(electrochemical migration)

(b) Electro migration (c) Stress migration

Reference (b) (c ): Tsuneo Ajiki: “reliability of semiconductor device”, Nikkagiren, 1988

Reaction Classification Phenomenon Failure part

Electrochemistry Ionic migration(electrochemical migration)

The metal ionizes, a metallic ion migretesby the electric field.

Between wiringfor PCBs

Electro migration Interaction of metallic atom and electron

Stress migration A metallic atom migrates by the mechanicalstress.

Themal migration A metallic atom migrates by the thermalstress.

Physical Aluminum wiringof semiconductor

Page 7: Ionic Migration on Printed-Circuit Boards

cathode(-)anode(+)

(a) Dendrite

M+ CAF

(b) CAF (Conductive Anodic Filament)

CAF

③ Resin

Copper electrode

cathode

anode

Dew or moisture

dendrite

moisture

0.5mm0.1mm

cathode(-)anode(+)

Solder dendrite on PCB Copper CAF in PCB

The forms of Ionic Migration

Phase1: metallic dissolutionPhase2: metallic ion migratePhase3: metallic deposition

Page 8: Ionic Migration on Printed-Circuit Boards

Metal Oxide-film (MOX)

Insulation layer

M

Mn+

Diffusion

M

M

MOx

Dissolution

MOxMOx

M(OH)x

Oxide- film metal

Deposit

OH-H+

Anode (+) Cathode (-)

M(OH)x

O2

H2O

Mn+

Dew or moisture

Ⅰ Anodic reaction

M→Mn++ne- ・・・・・・・・・・・・・・(1)

H2O→1/2O2+2H++2e-・・・・・・・・・(2)

M+H2O→MO+2H++2e-・・・・・・・ (3)

Ⅱ Cathodic reaction

Mn++ne-→M・・・・・・・・・・・・・・・・・(4)

O2+2H2O+4e-→4OH-・・・・・・・・・・(5)

2H2O+2e-→H2+2OH-・・・・・・・・・・(6)

2M++2OH-→M2O+H2O・・・・・・・・(7)

M2++2OH-→M(OH)2・・・・・・・・・・・(8)

Ⅲ Reaction between electrodes

2M++2OH-→M2O+H2O・・・・・・・・・(9)

M2++2OH-→M(OH)2・・・・・・・・・・・(10)

Reaction mechanism of Ionic Migration

Page 9: Ionic Migration on Printed-Circuit Boards

Acceleration factor of Ionic Migration

Factor Acceleration condition

Materials ( Fast) Ag>Cu>Pb>Sn-Pb Solder>Sn>Au( Slow)

Temperature High Temp

Humidity High humidity

Voltage High voltage

pH Acidity

Ionic impurities Halogen material (Chlorine, Bromine)Printed-cuircuitboard material Paper phenol > Glass epoxy > Polyimide > Ceramic

Page 10: Ionic Migration on Printed-Circuit Boards

Acceleration factor : materials (Pourbaix potential - pH diagrams)

immunitypassivity

corrosion

Tin (Sn)

0 7 142

1

0

-1

-20 7 14

Ag2O3 Ag 2O2

AgAg 2 O

Ag +

pH

0 7 142

1

0

-1

-20 7 14

Sn

SnO 2

SnH 4

Sn2+

SnO

32-

pHPo

tent

ial /

SH

E

Pote

ntia

l / S

HE

0 7 142

1

0

-1

-20 7 14

CuO

2-

Cu

CuOCu2+

pH

0 7 142

1

0

-1

-20 7 14

PbO 2

Pb2+

PbPbH2

PbO

pH

Pote

ntia

l / S

HE

Pote

ntia

l / S

HE

Silver (Ag)

Lead (Pb)

Copper (Cu)

Page 11: Ionic Migration on Printed-Circuit Boards

Acceleration factor : materials (Energy of oxide - film)

metal oxide-film △Gf o ( kJ/ mol )

SnO2 -515.47

Bi2O3 -496.64

Pb2O 3 -321.92

ZnO -321.92

SnO -257.32

PbO -189.33

CuO -127.19

Ag2O -10.75

Gibbs free energy change of metal oxide-film

Stability

Page 12: Ionic Migration on Printed-Circuit Boards

Acceleration factor : Humidity (water – film and corrosion)

0.1

1

10

100

20 40 60 80 100

relative hum i. (%)

amount of corrosio

(mg/dm2)

(a) Relative humidity and corrosion of metal(iron)

(b) Thickness of water-film and corrosion speed

thickness of w ater-film

corrossion spee

1mm~10μm~

wetdry waterwet

Reference (a): varon, w. : Trans, faraday soc., 23, 113(1927)(b):Tomashov, N D.: corros., 20, 7t(1964)

Page 13: Ionic Migration on Printed-Circuit Boards

Acceleration factor : Voltage

(a) Quantitative characteristics of elution: voltage and copper ions

(b)The relationship between applied voltage and migration occurrence

0

10

20

30

0 20 40 60

Tim e (sec)

Cu+/Cu2+decomposition amount

(mg/L)

5VD

2VD

3VD

1VD

Ta=25℃

0

100

200

300

0 1 2 3 4 5 6

applied voltage(V)

migration occurrence time(s

de-ionaizedw ater(pH6.6)

Ta=25℃

Page 14: Ionic Migration on Printed-Circuit Boards

Test methods

Page 15: Ionic Migration on Printed-Circuit Boards

Test standards for Insulation evaluation

Standard Test nam e Test conditions Applied voltageM easurem aentVoltage

IPC -TM -650-2.6.13

Assessm ent ofSusceptibility toM etallic DendriticG rowth UncoatedPrinted W iring

25℃De-ionized water:

60m l

0~20V/DC(M AX.15m A)

ANSI-J-STD-004

Requirem ent forSoldering Fluxes

85℃, 85%RH,168H

50V / DC 100V / DC

40℃, 90-95%RH,1000H

45 to 50V /DC

100V / DC

85℃, 85 to 90%RH,1000H

45 to 50V /DC

100V / DC

35℃, 85 to 93%RH,4days

100 V / DC

50℃, 85 to 93%RH,7days

100V / DC

Decided inconsultationwith purchaser

M oisture andInsurationResistance ofprinted boards

IPC -TM 650-2.6.3

JIS-Z-3284 Solder Paste

Page 16: Ionic Migration on Printed-Circuit Boards

Migration test pattern on PCB

(a) IPC test pattern (b) IPC multi-purpose test board

Page 17: Ionic Migration on Printed-Circuit Boards

Test methods : Simple migration test

(a) Water drop test

PCB

ElectrodeDe-ionized water

A

Power supply

Current meter

Resistor = 10kΩ

(b) Filter paper test

Resin

ElectrodeFilter paper

moisture

Resistor = 10kΩ

APower supply

Page 18: Ionic Migration on Printed-Circuit Boards

Test methods : Environmental migration test

Temperature / humidity chamber

HAST

Dew condensation cyclic chamber

Temperature / humidity cyclic chamber

(a) Evaluation by absorption (b) Evaluation by dew

Page 19: Ionic Migration on Printed-Circuit Boards

(1) specimen

(2) High Temp/High Humi. test

(3) Insulation resistance meter (HP4329)

(4) Insulation resistance meter (HP4339)

measurement method : Manual measurement

Page 20: Ionic Migration on Printed-Circuit Boards

Insulation resistance test of PCBs

Page 21: Ionic Migration on Printed-Circuit Boards

Results of Water Drop test (Cu and Ag, 2V/DC bias)

cathode(-)anode(+)

External appearance after testing

Copper migration

Silver migration

0.0

0.2

0 .4

0 .6

0 .8

1 .0

0 100 200 300

T im e (s e c )

Cu

rre

nt

(mA

)

0.0

0.2

0.4

0.6

0.8

1.0

0 100 200 300

Tim e (sec)

Current (mA)

Change in currentCopper (Cu)

Silver (Ag)

Page 22: Ionic Migration on Printed-Circuit Boards

Surface insulation resistance (SIR) test for PCB

(a) Test condition

(b) Specimen

Grass-epoxy PCB

40℃,87%60℃,87%85℃,85%

1000HrGlass-cloth epoxy (FR-4)

IPC-B comb pattern(Gap=0.3mm)

Applied voltage 50V / DCMesurement

intervals 1 Hr

Test conditions

PCB material

Ionic migration evaluation system (ESPEC)

Wiring situation

Page 23: Ionic Migration on Printed-Circuit Boards

Results of SIR test (Copper pattern PCB, 50V/DC bias)

Change in insulation resistance Change in absorption characteristic

1.E+09

1.E+10

1.E+11

1.E+12

1.E+13

0 200 400 600 800 1000

Tim e(hr)

Resistance(Ω

40℃87%RH

60℃, 87%RH

85℃,85%RH

0.01

0.1

1

0 200 400 600 800 1000

Tim e(hr)humidity absorption weight

40℃87%RH

60℃87%RH

85℃85%RH

Page 24: Ionic Migration on Printed-Circuit Boards

Results of SIR test (Copper pattern PCB, 50V/DC bias)

40℃, 87% 60℃, 87% 85℃, 85%

External appearance after testing (2000Hour)

No migration No migration No migration

cathode(-)

anode(+)

Page 25: Ionic Migration on Printed-Circuit Boards

Example : Results of SIR test (Paper phenolic PCB)

85℃, 85%, 50V/DC, 2000Hr

External appearance after testing

110℃, 85%, 5V/DC, 300Hr

cathode(-)

anode(+)

CAF growth CAF growth

CAF

CAF

Page 26: Ionic Migration on Printed-Circuit Boards

1.0E+06

1.0E+07

1.0E+08

1.0E+09

1.0E+10

1.0E+11

0 200 400 600 800 1000

Tim e (Hour)

Resistance (Ω

Example : Results of SIR test (Flux and Dew test)

Evaluation of soldering flux( 85℃85%,50V/DC bias )

Flux A

Flux B

Flux CMigration growth

Dew cyclic test( 5℃/25℃90%,5V/DC bias )

1E+06

1E+08

1E+10

1E+12

1E+14

1E+16

0 20 40 60 80 100

Tim e (C ycle)

Resistance(Ω

Migration growth

Page 27: Ionic Migration on Printed-Circuit Boards

IPC-9201 (Surface Insulation Resistance Handbook)

)1122()]

21

11(exp[)]

21

11)(

REaexp[(12

VAVA

RHRHb

TTtt ×−×−×=

The Institute of Electrical Engineers of Japan (Report No. 772)

Ea = Activation energy ( Glass Epoxy PCB = 1eV)

R = Boltzmann’s Constant (8.63×10-5 eV/k)

T1, RH1, V1, D1, t2 = Practice use Temp., Humi., Voltage, Electrode Distance, Time

T2, RH2, V2 , D2, t1= Accreted test Temp., Humi., Voltage, Electrode Distance, Time

B, A1,A2, r, n, m = constant ( r=3, n=2)

mDDn

VVr

RHRH

TTREaAF ⎟

⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛ −=

21

12

12)

21

11(exp

Life estimated formula of insulation failure

Page 28: Ionic Migration on Printed-Circuit Boards

・Ea = Activation energy

(Glass Epoxy PCB = 1eV)

・R = Boltzmann’s Constant (8.63×10-5 eV/k)

・ Practice Temp.(T1)=45℃,Humi.(RH1)=85%

Voltage(V1)= 5V

・Test Temp(T2)=85℃,Humi.(RH2)=85%

Voltage(V2)=5V, 2000Hr

・r, n = constant ( r=3, n=2)

・AF1= exp[1.16×104×(1/318-1/358)]

= exp(4.1) = 59

・AF2 = (85/85)3 = 1

・AF3 = (5/5)2 = 1

※AF=59×1×1 = 59

Life Time = 2000Hr×AF

= 118000Hr = 13.4 years

mDDn

VVr

RHRH

TTREaAF ⎟

⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛×⎟

⎠⎞

⎜⎝⎛ −=

21

12

12)

21

11(exp

Life estimated formula of insulation failure

Reference: The Institute of Electrical Engineers of Japan (Report No. 772)

Page 29: Ionic Migration on Printed-Circuit Boards

Solder alloy migration

Page 30: Ionic Migration on Printed-Circuit Boards

(a) SEM image of solder surfaces

(b) Sn-0.75Cu

(a) Sn-3.5Ag (c) Sn-9Zn

Ag3Sn

Zn

Pb

20μm

CuxSnx

(b) Standard potential of solder composition materials

Factors of Solder alloy migration

(d) Sn-37Pb

E ゚(V vs. SHE)

base Zinc Zn2++2e-=Zn -0.763

Tin Sn2++2e-=Sn -0.138

Lead Pb2++2e-=Pb -0.126

Hydorogen 2H++2e-=H2 0.000

Bismath Bi3++3e-=Bi 0.215

Copper Cu2++2e-=Cu 0.337

Nobl e Silber Ag++e-=Ag 0.779

m aterials Reaction

Page 31: Ionic Migration on Printed-Circuit Boards

-1000

-800

-600

-400

-200

0

200

400

Pote

ntia

l / m

V vs

. SC

Electrochemical characteristics (Static characteristic)

Zn

Sn-37Pb

Sn-0.75Cu

Sn-58Bi

Cu

Ag

Sn-3.5AgSn

base (active)

Noble

Pb

Sn-9Zn

Rest potential of each type solder (in 0.1M KO3 aqueous solution)

Sn-8Zn-3BiSn-3Ag-0.5Cu

Page 32: Ionic Migration on Printed-Circuit Boards

Electrochemical characteristics (Dynamic characteristic)

Simple metals Solder Alloys

Sn-3.5Ag Sn

Zn

Sn-9Zn

Sn-37Pb

-2000 -1500 -1000 -500 0 -2000 -1500 -1000 -500 0 500

Potential / mV vs. SCE

Cur

rent

den

sity

/m

Acm

-2

Pb

500- 10

0

10

20

30

- 10

0

10

20

30

- 10

0

10

20

30

Current - potential curves = dissolution characteristic (in 0.1M KO3 aqueous solution)

Dissolution is easy

Page 33: Ionic Migration on Printed-Circuit Boards

0

5

10

15

20

0 25 50 75 100

0

0.1

0.2

0.3

0.4

0.5

0 100 200 300 400

20 sec 90 sec 123 sec

Water Drop test - 1(5V/DC bias)

250 sec 332 sec 350 secSn-Zn

Time (sec)

Cur

rent

(mA

)Sn-Pb 10 sec 20 sec 30 sec

Change in current

0

0.01

0.02

0.03

0.04

0.05

0 100 200 300 400

Sn-Ag

Process of migration depositsSn-Pb

Sn-Ag

Sn-Zn

cathode(-)

Page 34: Ionic Migration on Printed-Circuit Boards

A

B

C

200μm

Water Drop test - 2(SEM analysis)

SEM image of migration deposits

Sn-Zn

Sn-Pb

Sn-Ag cathode(-)

anode(+)

Compositional analysis of migration deposits

A

B

C

A

B

C

solder Point A Point B Point C

Sn = 63% Sn = 46% Sn = 47%

Pb = 37% Pb = 54% Pb = 53%

Sn = 100% Sn = 100% Sn = 100%

Ag = 0% Ag = 0% Ag = 0%

Sn = 98% Sn = 93% Sn = 97%

Zn = 2% Zn = 7% Zn = 3%

Sn-Pb

Sn-Ag

Sn-Zn

Page 35: Ionic Migration on Printed-Circuit Boards

1.0E+08

1.0E+09

1.0E+10

1.0E+11

0 250 500 750 1000

1.0E+08

1.0E+09

1.0E+10

1.0E+11

0 250 500 750 1000

Sn-Ag-Cu

Sn-37Pb

Solder migration

Breakdown

Surface insulation resistance test (50V/DC bias, 85℃85%)

No migration

Migration occurringTime (Hour)

Res

ista

nce

(Ω)

Time (Hour)

Res

ista

nce

(Ω)

Change in insulation resistance

Page 36: Ionic Migration on Printed-Circuit Boards

Evaluation using mounting PCBs(THB Testing, 5/DC bias,80℃90%)

Terminal of QFP (spacing:0.5mm)

Chip capacitance (spacing:0.8mm)

No migration

No migrationSn-Ag-Cu solder

5V GND

5V GND

END