A Fracture Mechanics Based Approach for Adhesion Testing in the PV Module Laminate

Nick Bosco and Sarah Kurtz National Renewable Energy Laboratory

Reinhold H. Dauskardt Stanford University

NREL is a naAonal laboratory of the U.S. Department  of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

 

 

 

 

 

overview

§ Short review of common PV adhesion tests

§ Introduction of the fracture mechanics based approach to adhesion testing

§ Applications of this approach to PV laminate materials at both the coupon and module level

§ Extension of these measurements for lifetime prediction of adhesive systems

§ Review and Direction

2

limitations of common adhesion tests lap shear

peel test

3

fracture mechanics

fracture dWsGc = toughness dA

complianc δC = e P energy release rate toughness adhesion debond energy

G = P2

2b

dC

da

4

figures inspired by: T.L. Anderson, “Fracture Mechanics” CRC Press, Boca Raton, FL 1995

sample modification

peel test

cantilever beam

5

single cantilever beam glass/EVA

6

 

 

 

single cantilever beam 0% silane

§ Load reversals to measure compliance with crack extension C ∝ a3

§ produce linear fit of compliance with crack extension

§ Evaluate toughness at each crack length

P2 dC G = = 2b da

Pn 2

2b 3man

2C = ma3 + b

dC

da = 3ma2

7

single cantilever beam, coupon level 0% silane

0.8 mm Ti beam 1.5 mm Ti beam

2mm 2mm

Calculate fracture toughness for each crack extension

8

single cantilever beam, module level 0% silane

9

single cantilever beam, EVA

10

20 30

single cantilever beam, backsheet Debond Energy decreased with aging temperature and RH

1200

Deb

ondEne

rgy,

G(J/m

2 )

c

40 50 60 70 80 90

Ageing Ame =1000 hrs

80% RH

0% RH

1000

800

600

400

200

0 Unaged

A geing T empera ture (°C )

11

12  

single cantilever beam, backsheet

0 200 400 600 800 10000

200

400

600

800

1000

1200

Unaged

 

 

Deb

ond  Ene

rgy,  G

c  (J/m

2 )

A ge ing  T rea tment  D ura tion  (hrs )

Debond Energy decreased with aging duration

85°C,  85%RH  

13  

double cantilever beam (DCB) thin metal film on glass

δ

P

14  

corner adhesion test

G =12PP

2

E 2 tan θ2!

"#$

%&

!

"#

$

%&

2

h3

G =P2

2bdCda

b = 2a tan θ2!

"#$

%&

§  Compliance becomes independent of crack length

§  Crack will extend at a constant,

critical load

debond experiment

0 500 10000

10

20

30

40

 

 

Deb

ondi

ng L

oad,

Pc (N

)Actuator Displacement, Δ (µm)

15  

corner adhesion test, coupon level

Corner

SCB

16  

corner adhesion test, coupon level

100 110 1200

30

60

90

120

150

 

 Deb

ond  Ene

rgy  (J/m

2 )

C uring  T empera ture  (°C )

PV-86, 300 µm

PV-86, 100 µm

PV-5414, 500 µm

Ionomers t = 30 min

Effect of Curing on Ionomer Bonding Strength

17  

corner adhesion test, module level

18  

corner adhesion test, module level

applied to module cell applied to module backsheet

19  

corner adhesion test, module level

1

10

100

1000

 

 

Deb

ond  Ene

rgy  (J/m

2 )

New  Module   10  years  in  field  

Gc=  2.8  J/m2  Debo

nd  Ene

rgy,  G

c(J/m

2 )  

debond energy decreased 500-fold after 10 years in

Florida

20  

subcritical crack growth

dadt= −

dPdt

"

#$

%

&'ma3 + b3Pma2

Gn =P tn( )2

2b3ma tn( )2

21  

subcritical crack growth V-G plots

This is the “adhesive strength”

This is the “adhesive threshold”, or a stress at which cracks will not propagate. A reliable design will “live” here.

22  

subcritical crack growth, SCB Glass/ EVA

23  

subcritical crack growth, SCB Glass/ EVA debond growth kinetics

0 200 400 600 800 1000 1200 140010 -­‐8

10 -­‐7

10 -­‐6

10 -­‐5

10 -­‐4

10 -­‐3

 

 

Deb

ond  Growth  R

ate,  da/dt  (m/s)

D ebond  D riving -­‐F orce ,  G  (J /m 2)

RH=10%  

25%  

40%  

55%  70%  

T=30oC  

debond growth of EVA encapsulant interfaces is controlled by the viscoelastic processes that are affected by water through the plasticization of the debond tip

EVA dc

rp

Viscoelastic Relaxation and Cavitation

dadt=π8

δcεy δcεyE1( )

1 n G1 n10

Ca T+γRH−Tg−dry( )Cb+ T+γRH−Tg−dry( )

24  

subcritical crack growth, SCB Backsheet debond kinetics and lifetime prediction

0.1

1

10

100

1000

10°C

40°C

30°C

20°C

10              20              30              40              50              60

 

 

Bac

kshe

et  Life

time  (yea

rs)

Initia l  D efec t  S iz e ,  di(µm)

da = dadt0

t∫ai

a f∫ dt

dadt= kG1 n10

Ca T+γRH−Tg−dry( )Cb+ T+γRH−Tg−dry( )

25  

direction

§  Ongoing NREL scientific work is focused on applying the FM method to characterization for all PV adhesives

§  We will develop protocols for applying this technique to all relevant material systems

§  This work will provide the scientific basis for incorporating these techniques into future revisions of international standards

26  

limitations of common adhesion tests

Sample Properties

§  Stiffness §  Electrical Resistance §  Strength

Material Properties

§  Elastic Modulus §  Electrical Resistivity §  Toughness

27  

review

§  FM based adhesion tests measure a quantitative material property

§  Methods can be applied at both the coupon and module level and to all interfaces of the PV laminate

§  Tests may be developed to be straight forward using common mechanical test equipment

§  Subcritical measurements allow modeling of adhesive degradation mechanisms and ultimately provide a lifetime prediction tool