1

Non Contact Electrical Characterization of PV Films

Review of conductivity/dielectric/insulation resistance testing • Direct Current (DC) Insulation Resistance Standard Test

(IEC 62788-1-2, 2014 developed for PV insulators) • Alternating Current (AC) Insulation Resistance Test

(ASTM D149 modified for printed circuit boards dielectrics) • Brief intro to polarization and conduction processes in dielectrics • real and imaginary AC conductivity • correlation between DC and AC conductivity

Non-contact AC conductivity measurement using a 7 GHz resonant cavity

• AC conductivity of PET – effect of exposure to environmental conditions • Estimation of the corresponding insulation resistance change

Jan Obrzut, NIST, MML, Div 642 (jan.obrzut@nist.gov)

Contributors: Xiaohong Gu, NIST, EL

3rdAtlas/NIST Workshop on Photovoltaic Materials Durability December 8 and 9, 2015

2

International Standards for PV modules

IEC 61215 (crystalline silicon), IEC 61646 (thin films)

Requirements for the design qualification and approval of terrestrial photovoltaic modules suitable for long-term operation in general open air climates

Insulation resistance passing criteria a) the degradation of maximum output power does not exceed the prescribed limit after

each test nor 8 % after each test sequence; b) no sample has exhibited any open circuit during the tests; c) there is no visual evidence of a major defect, as defined in Clause 7; d) the insulation test requirements are met after the tests;

R >= 400 MΩ if area < 0.1 m2, Dielectric withstanding voltage 1000 V for 1min R *area >= 40 MΩ m2 at 500 V if area > 0.1 m2

e) the wet leakage current test requirements are met at the beginning and the end of each sequence and after the damp heat test;

f) specific requirements of the individual tests are met.

3

HV DC Insulation resistance test ( IEC 62788-1-2, 2014)

S

sCRt

i Rve

RVti Si +≈ − /0)(

i(t) V0

Ri

RS CS

is RR >>

Rs Cs

Ri V0

itest =V0/Rs ≈ 2pA, Rs = 103 V/(2 x 10-12 A) = 5 x 1014 Ω ρ = Rs As / ts = 2.2 x 1017 Ωcm

Rs-field (0.1 m2) = 2.2 x 1017 Ωcm *0.046 cm /103 cm2 ≈ 1013Ω ρmin (Rs-min-field = 4 x 108 Ω, ) ≈ 8.6 x 1012 Ωcm

(V0/Rs )max-test ≈ 5 x 10-8 A, will get there in about 1.3 105 cycles, if linear conditions stay

0

200

400

600

800

1000

1200

-2

-1

0

1

2

3

4

0 1 2 3 4 5

Appl

ied

Volta

ge (V

)

Curr

ent (

pA)

Time (h)

Example Data

Current (pA) Voltage (V)

SRV0

As =20.5 cm2

ts = 0.046 cm

Step voltage stimulus – current decay response in time domain

transient specimen current

4

Pros: • Simple instrumentation, easy visualization • Widely accepted • Large data base of performance (failure rate) in commercial applications

(oldest standards , R. Bartnikas, 1982 )

HV Direct Current Insulation Resistance Test

Cons: • Long testing (electrification) time to read the steady state leakage current is-∝ =

V0/Rs • Arbitrary acceptance criteria, often based on historic performance (R> 400 MΩ) • All signals are transient until the steady state conditions are reached, then vs = V0 • Test results are difficult to link with the physical mechanism of aging and reliability

projection (the current decay and kinetics of charging are consequence of several processes acting simultaneously: dipolar polarization, charge transport, space charge )

Step voltage stimulus – current decay response in time domain

5

Alternating Current (AC) Insulation Resistance Test

)()i( 0 ϕω += tRVt

isin

0 5 10 15 20

-600-400-200

0200400600

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

V 0 (V)

t (ms)

V0 VRi

V Ri (V

)

)1(i −= ∗

∗∗

RiS R

VVZ

o5892 .−≈= τ∆πϕ f

Sinusoidal voltage stimulus – alternating current response at single frequency

in time domain )()v( 0 tVt ωsin=

Rs

∗SZ -complex impedance

sC

Ri ∗RiV

∗V

In frequency domain ))((0 ϕω +=∗ tjVV exp

-89.3

-89.2

-89.1

-89.0

0 500 1000 1500 2000 250044.5M

44.6M

44.7M

44.8M |ZS|

|ZS|

(Ω)

Voltage (V)

Phas

e φ (d

egree

s)

φ

ϕ= -90° real (ideal) capacitance 0> ϕ >-90° complex cap with loss (ZS*) ϕ = 0 real impedance , ZS = Rs

Vmax

Withstanding voltage Vmax ≈ 1.2 kV Rs = 5.2 ×1011 Ω

J. Obrzut, IEEE Trans. Instr Meas., 54, 1570 (2005)

f= 50 Hz

∆τ ≈ 5 ms

( ))(|111

S|sin

Zϕ−≈

∗SR

6

Waveform Measurements in Time and Frequency Domain

• Time domain - transient response, visualization • Frequency domain - steady state phasor

transforms are convenient for calculating materials property from complex conductivity

• Non-linear response – harmonics early indicator of dielectric-breakdown

τ

V0 - amplitude ϕ =ωτ - phase

v(t)

t

f

Freq. domain V* = V0 e i(ωt+ϕ)

Time domain v(t) = V0 sin (ωt+φ)

∗SZ ∗

SCRs

Ri ∗RiV

∗∗∗ +== SSSS jR C/Y/Z ω11

σσσ ′′+′== ∗ jA

t

SZ*

∗+= εωεσσ 00 j*

rεωεσσ ′′+= 00'

rj εωεσ ′= 0"

Complex conductivity σ* in frequency domain

∗V

J. Obrzut, IEEE Trans. Instr Meas., 54, 1570 (2005)

DCi ciε ′′i

DCiε ′′i

ci

Rii + +=

Rii

7

0.003 0.004 0.00510-9

10-8

10-7

10-6

10-5

10-4

1011021031041051061071081091010101110-10

10-9

10-8

10-7

10-6

10-5

10-4

193 K

333 K

1/T (K-1)

Cond

uctiv

ity (S

/m)

σ' (S

/m)

frequency (Hz)

σ0

σ' at 50 MHz

σ0

σ0 ≤ σ′ ; σ′(ω) → σ0 ω → 0 = σDC ; universal scaling law σ′(ω) − σ0 ∼ ω n :

00 )( ωεσσε /'−=′′

Example of AC conductivity ( σ′ ) and dielectric relaxation (ε″) scaling with frequency and temperature

εωεσσ ′′+= 00'

Change in Ea-σo , Ea-σ’, Ea-rlx, ε″max , fmax and exponent n can be used to determine physical mechanism of aging

J. Obrzut, Phys.Rev. B, 76, 195420 (2007), Phys.Rev. B, 80, 195211 (2009)

Ea-rlx=9 kJ/mol

Ea-σo=0.24 eV

Ea-σ’=0.1 eV

fmax

Dielectric relaxation / Arhenius plot

8

Couplers rotated 89° (cross-polarized) create impedance termination

ff

Q peak

∆=

cavity

couplers

specimen

E

H

Ey

Hx Hz

Conductive loss causes the resonant peak to broaden and decrease the quality factor Q.

c

sc

fff −Frequency shift in position of the resonant

peak is proportional to the specimen dielectric constant.

The specimen conductivity and permittivity can be determined from the measured change in Q-factor and frequency shift.

Non Contact Measurement of AC Conductivity by the Microwave Resonant Cavity Technique

partial insertion: Vx = w t hx

IEC TS 62607-6-4 “Graphene – Surface conductance measurement using resonant cavity” J. Obrzut 2015

9

020

4060

80100

120140

0.00.20.40.60.81.0

02 4 6 8 10

020

4060

80100

120140

0.00.20.40.60.81.0

02 4 6 8 10

020

4060

80100

120140

0.00.20.40.60.81.0

02 4 6 8 10

7.0x109 8.0x109 9.0x109 1.0x1010-90

-80

-70

-60

-50

-40

-30

-20 TE106TE105TE104TE103

S 21(dB

)

frequency (Hz)

TE102

0 20 40 60 80100120

140

0.00.20.40.60.81.0

0246810

0 20 40 60 80100120

140

0.00.20.40.60.81.0

0246810

x (mm)

z (mm)(E

y/E0)2

7.30x109 7.32x109

-35

-30

-25

TE103

S 21(d

B)

frequency (Hz)

fo103 = 7.3191125 GHz ± 50 kHz, frequency resolution 10-6 !

small changes in the specimen conductivity can be easily detected

Non Contact Measurement of Conductivity by the Microwave Resonant Cavity Technique

10

-26

-25

-24

-23

-22

-21

7.42x109 7.43x109 7.44x109-26

-25

-24

-23

-22

-21

S 21 (d

B)

Frequency (Hz)

Exposed PET f0

increasing specimen volume

Fresh PET

partial insertion: Vx = w t hx

0.0 0.2 0.4 0.6 0.8 1.0-1x10-5

01x10-52x10-53x10-54x10-55x10-56x10-57x10-5 0 days 54 µS/cm

184 days 60 µS/cm 392 days 64 µS/cm 510 days 66 µS/cm fits

1/Q

x-1/

Qo

k*Vx (cm/S)

σ,v= slope of the line

qxvx

bVVfQQ

2211

0000−=−

πεσ

Effect of environmental exposure on MW conductivity of PET (ambient RH, T, sunlight exposure)

• MW Conductivity increases with duration of the environmental stress

• DC conductivity too small to measure

Specimen size: 5 mm × 295 µm × 10 mm

(Eq. 1)

IEC TS 62607-6-4 “Graphene – Surface conductance measurement using resonant cavity” J. Obrzut 2015

11

Days σ (µS/cm) στ/σ0 0 54 1 184 60 1.11 392 63 1.16 510 66 1.22 τfailure 1.4 S/cm 2.5 x104

Projected degradation of HV DC resistivity

(Rs-min-field = 4 x 108 Ω, ) ≈ 8.6 x 1012 Ωcm

Effect of environmental exposure on MW conductivity of PET

HV DC Insulation resistance test (per IEC 62788-1-2, 2014)

Days ρ (Ω cm) 0 2.2 × 1017 184 2.0 × 1017 392 1.9 × 1017

510 1.8 × 1017 τfailure 8.6 × 1012

AC (7GHz) conductivity increase

There is no indication that the environmental exposure would compromise

HV DC insulation resistance minimum requirements

0 2 4 6 8 10 12

-2

-1

0

1

2

ln(ln

(σt/σ

0))

ln(t) (days)

ρ0/ρmin= 2.5× 104

25 years

50 years

12

MW Conductivity of PET, after UV and Temp accelerated stress test

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

0.05.0x10-6

1.0x10-5

1.5x10-5

2.0x10-5

2.5x10-5

3.0x10-5

3.5x10-5

4.0x10-5

UV, 85oC, 5 % RH UV, 85oC fresh PET fit

1/Q

x-1/

Qo

k*Vx (cm/S)

22 Days

Conductivity of PET samples increases after the stress, but at low RH level the effect of UV and Temp is not that significant

Measurement: Plots of Quality factor change in conductivity notation (Eq. 1)

Conductivity, σ= slope of the plot

Conductivity, σ= slope of the plot

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7-5.0x10-6

0.05.0x10-61.0x10-51.5x10-52.0x10-52.5x10-53.0x10-53.5x10-54.0x10-5 UV, 85 oC, 5% RH

UV, 85 oC Control σ=60 µS/cm fit

1/Q

x-1/

Qo

k*Vx (cm/S)

11 Days

13

MW Conductivity of PET, after exposure to UV, Temp and 60% RH

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

0.01.0x10-5

2.0x10-5

3.0x10-5

4.0x10-5

5.0x10-5

6.0x10-5

7.0x10-5

8.0x10-5

30 days + TS, σ= 164 µS/cm 30 days, σ= 130 µS/cm 22 days, σ= 75 µS/cm 11 days, σ= 65 µS/cm Control, σ= 60 µS/cm fits

1/Q

x-1/

Qo

k*Vx (cm/S)

UV, 85oC, 60 % RH,

0 2 4 6 8 10-3

-2

-1

0

1

2

3

ln(ln

(σt/σ

0))ln(t) (days)

ρ0/ρmin 25 years

Tensile tested

UV, T, RH

Under UV, Temp stress conditions Humidity dramatically accelerate loss of insulation resistance. Mechanical elongation creates additional conducting paths.

Ambient sunlight

50 days

14

Tensile test results

0.0 %

0.48 %

0.65 %

1.07 %

1.78 %

17.60 %

(a) (e)

(f) (b)

(c)

(d)

Crack density increases with RH

15

• Direct Current (DC) High Voltage Insulation Resistance Test: step voltage – current decay in time domain easy to visualize, commonly used

• Alternating Current (AC) Insulation Resistance Test phasor transform in frequency domain fast, physical mechanism of charge transport at high frequencies eliminates ambiguity with ionic current (redox process)

• Demonstrated non-contact resonant cavity test method for conductivity of PET samples. The method operates at 7 GHz and is sensitive to aging effects caused by the UV, T, RH ambient and accelerated stress. No direct correlation with the life test.

• sensitive, fast, non-invasive, small specimens (New IEC std developed at NIST).

0 ; )(0 == ′′∞ ετ iiiVR cR ,/

)( )()0()( ωωωω ε CRAC iiii ++== ′′

ACDC RR >

Summary

Thank you!

( ))(|111

S|sin

Zϕ−≈

∗SR

3rdAtlas/NIST Workshop on Photovoltaic Materials Durability December 8 and 9, 2015