1Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Through-Plane & In-Plane Conductivity of Polymer Electrolyte Membranes

Scribner Associates, Inc.www.scribner.comSeptember 2010

2Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

1

10

100

0 10 20 30 40 50 60 70 80 90 100 110Relative Humidity (%)

Con

duct

ivity

(mS

/cm

)

This work, Through-planeMarechal (2007), Through-planeMa (2006), Through-planeMa (2006), In-planeSone (1996), In-planeZawodzinski (1991), In-planeBekkTech (unpublished), In-planeKreuer (1997), In-planeLee (2005), In-planeSummer (1998), In-plane, RT

Nafion® 117, 30 oC

~ 10x !

Large discrepancy in reported membrane conductivity data highlights need for accurate, robust measurement methods

3Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Objective: Develop an accurate & reliable test apparatus & method for through-plane membrane resistance & conductivity measurements

Key desirable featuresUses bare (non-catalyzed) membrane

relevant thicknesses – 10 to 200 µmOperate over a wide range of conditions

• 30 to > 120 oC• dry to > 95% RH• 1 to 3 atma

Rapid ~ 15 min per test conditionRobust - accurate, repeatable and

reliable

A L

]/[ cmSAR

L

membranemembrane ⋅

=σ

4Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Membrane Test System MTS 740

Mass flow controllers (MFC) for wet-dry gas mixing

Cell head w/ integrated electrodes, clamping mechanism, lead connector

Test chamber with in-situ dew point & sample temp

Temp controls –test chamber, humidifier, line

BackpressureregulatorCondensate

collection tank

E-stopbuttonAuto-water

humidifier

U.S. Patent No. 7,652,479

5Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Wet-dry gas mixing for rapid RH cycling in both directions

MFC

MFC

TC

N2H2BP

VentWatertrap

In-situ dew point &temperature probes

Condenser

6Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Wet-dry gas mixing for controlled, rapid RH cyclingDifference

from Nominal, % RH

Nominal% RH

30 °CN = 15

80°CN = 11

20 0.5 -0.340 0.7 -0.560 -0.2 -0.880 -1.4 -0.490 -1.5 0.395 -2.2 -0.3

40

50

60

70

80

90

12:00 12:30 13:00 13:30 14:00 14:30 15:00 15:30

Dew

Poi

nt o

r Tem

pera

ture

, o C

0

20

40

60

80

100

Rel

ativ

e H

umid

ity, %

Dew PointSpecimen TRH

Specimen T

DP

RH

Repeatable, reproducible and stable T, dew point & RH ±2% from 20% to 95% RH

Rapid RH cycling time-efficient testing over wide RH range

Dew point to 120 oC, sample to 150 oC

7Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Comparing through-plane & in-plane conductivityof Nafion® NR-212

1

10

100

1000

10 20 30 40 50 60 70 80 90 100

Cond

uctiv

ity (m

S/cm

)

Relative Humidity (%)

30C, Through-Plane (SAI)30C, Through-Plane (SAI)120C, Through-Plane (SAI)30C, In-Plane (BekkTech)80C, In-Plane (BekkTech)120C, In-Plane (BekkTech)

σin-plane ≅ σthrough-plane for dispersion cast Nafion®

Nafion NR-212120 oC

80 oC

30 oC

8Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Is the conductivity of Nafion® isotropic? … No consensus in published literature

Yes, it is isotropic, σ׀׀:σ⊥≅ 1 [Nouel; Silva] This work for NR-212

No, it is anisotropic σ׀׀:σ⊥= 3.6 [Gardner] σ׀׀:σ⊥ = 2.5 – 5 (with pressure) [Ma] σ׀׀:σ⊥= 1.8 – 5 [Casciola]

Discrepancy due Different water content (λ) Extruded (N11X) vs. dispersion

cast (NR-21X)

Gardner et. al., J. Electroanal Chem 449 209 (1998)Ma et. al., JES 153 A2274 (2006)Casciola et. al., J. Power Sources 162 141 (2006)Nouel, Fedkiw, Electrochimica Acta 43 2381 (1998)Silva, et. al., J. Power Sources 134 18 (2004)

TDC = σ׀׀NDC = σ⊥N117

% RH 30 °C 80 °C 120 °C20 1.19 0.92 1.2040 0.95 1.02 1.1560 1.15 1.02 0.9680 0.99 0.96 0.8790 0.93 0.97 0.8795 0.93 106 0.84

σ׀׀:σ⊥ ≅ 1, NR-212, this work

Silva, JPS ‘04

σ׀׀ = in-plane, σ⊥ = through-plane

9Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Effective conductivity (σeff) of membrane with phases of unequal conductivity, e.g., ionomer-impregnated non-

conductive porous support

σeff, in-plane > σeff, through-plane for supported membrane σeff, in-plane : σeff, through-plane is a maximum for f = 0.5 σeff, in-plane : σeff, through-plane → 1 as f → 0 or 1 σeff, in-plane : σeff, through-plane increases as σ1 : σ2 → 0 or >> 1

f·Z

(1-f )·Z

σ1

σ2

In-Plane

Through-Plane ( ) 21 1 σσσ ⋅−+⋅= ffplane-in ,eff

( ) 21

21

1 σσσσσ

⋅+⋅−⋅

=ffplane-through ,eff

f = fractional thickness of phase 1

10Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Comparing through-plane & in-plane conductivity (σ) of PFSA-based membranes with inert support GORE-SELECT®σeff, in-plane > σeff, through-planeσeff, in-plane : σeff, through-plane → 1 as f → 0 or 1 Ratio is greater for thin membrane with same support thickness

Thickness

18 µm 1.53 ± 0.16

35 µm 1.11 ± 0.10

plane-through ,

plane-in ,

eff

eff

σσ

1

10

100

10 20 30 40 50 60 70 80 90 100

Con

duct

ivity

(mS

/cm

)

Relative Humidity (%)

80 oC

Gore SELECT® 18um Through-PlaneGore SELECT® 35um Through-PlaneGore SELECT® 18um In-PlaneGore SELECT® 35um In-Plane

1. Conductivity based on thickness measured at ambient temperature & RH2. Through-plane resistance corrected for non-membrane ohmic resistance

GORE-SELECT, GORE and designs are trademarks of W. L. Gore & Associates, Inc.

11Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Conclusions – Membrane Test System MTS 740 Through-plane resistance & conductivity test system developedBare membrane – rapid, lower cost assessment vs. MEA / fuel cell

testingRepeatable, accurate control of environmental conditions: cell to

150 oC, humidifier to 120 oC, dry to >95% RHRobust method – repeatable and accurate

Correction for non-membrane ohmic resistance contributions is important, especially for thin membranes with low resistance / high conductivity

Dispersion cast Nafion® NR-212 though-plane conductivity is the same as in-plane

Differentiate in-plane and through-plane conductivity for anisotropic material, e.g., GORE-SELECT® supported membrane

GORE-SELECT, GORE and designs are trademarks of W. L. Gore & Associates, Inc.

12Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Supporting Information

Electrode design

Test procedure

Analysis Procedure

Determination of Cell Resistance

13Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Membrane

Pt Source Electrode #1

Pt Source Electrode #2

Dimensions in mm

4-Electrode, “Offset” Electrode Design

Pt V-sense #1

Pt V-sense #2

GDE #2

GDE #1

Overlap Area BottomTop

Top View

U.S. Patent No. 7,652,479

Side View

14Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Offset electrode geometry for 4-probe conductivity measurement of thin film electrolytes

VS2 = 0.5 V

X, mm

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

-0.5

-0.4 -0.3

-0.2 -0.1

0.0

0.1

0.2

0.3 0.4

0.5

E, V

100 µm

Y-axis is expanded 10x relative to X-axis

Model condition: constant 1 V between source electrodes

Source Electrode #1

X, mm

Source Electrode #2

VS1 = -0.5 V

U.S. Patent No. 7,652,479

15Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Procedure – Pre-test

As-received membrane, stored at ambient conditions

32 mm x 10 mm sample Measure “dry” membrane thickness Mean of 5 locations, 3x measurements/location Low load, high accuracy gage

Cell Assembly GDE (E-LAT) cut with jig Glue GDE to Pt electrode with carbon paste Load membrane between GDE-prepared plattens Compress ~ 2,200 kPa (325 psi) using spring loaded cell

head (dial gage)

16Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Procedure

Temperature series (oC): 80 30 120 Per temperature

Wet-up 2 hr @ 70% RH RH cycle: 70 20 90 95 %, 15 min step Impedance sweep after 15 min

~ 1 day/temperature, ~6 hr Gas: H2 or N2 Impedance Measurement

4-electrode, 4 terminal Solartron 1260 FRA (standalone) / ZPlot®

10 MHz – 1 Hz, 10 mVAC, 0 VDC, 10 steps/dec (~ 2 min)

Temp, oC Total Dry Gas Flow, sccm Pressure, kPaa

30 500 10080 500 100120 500 230

17Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Post-test Procedure – EIS Analysis

R_hf R1

CPE1

R2

CPE2

0 1 2 3 4 5 610-2

10-1

100

101

102

103

|Z|,

ohm

20% RH40% RH60% RH80% RH95% RH

0 1 2 3

-1.5

-0.5

Z', ohm

Z'', o

hm

-50 50 150 250 350

-150

-50

Z', ohm

Z'', o

hm

20% RH40% RH60% RH80% RH95% RH↑RH

100 101 102 103 104 105 106

-100

-50

0

Frequency, Hz

Thet

a, d

eg.

Rhf

Increasing RH,decreasing Rhf

↑RH

20%40%

60%

95% 80%

18Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Post-test Procedure – EIS Analysis

R_b R1

CPE1

R2

CPE2

-5 0 5 10 15

-20

-15

-10

-5

0

Z'

Z''

G1 GDE 80C 06172008_001.ZFitResult

100 101 102 103 104 105 106 1070

5

10

15

Frequency (Hz)

|Z|

G1 GDE 80C 06172008_001.ZFitResult

100 101 102 103 104 105 106 107

-200

-150

-100

-50

0

Frequency (Hz)

thet

a

-1 0 1 2

-3

-2

-1

0

Z'

Z''

G1 GDE 80C 06172008_001.ZFitResult

RHF

Element Freedom Value Error Error %R_b Free(+) 0.068111 0.0011616 1.7055R1 Free(+) 12.92 0.087668 0.67854CPE1-T Free(+) 1.3459E-05 3.089E-07 2.2951CPE1-P Free(+) 0.83049 0.0017289 0.20818

RHF

19Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Through-plane Resistance & Conductivity

Through-plane resistance includes non-membrane ohmic contributions, Rcell

Typically work in area specific resistance, ASR

Accounting for the cell ASR(T,RH) gives the membrane resistance

Note that all are a f (T, RH)

The challenge: need ASRcell (T,RH)

][ 2effectiveHFduncorrecte cmΩARASR −⋅=

]cmΩ[),(),(),( 2−−= RHTASRRHTASRRHTASR cellduncorrectemembrane

interfaceΩ,contactΩ,electrodeΩ,cell RRRR ,Resistance Cell ++=

20Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Determine ASRcell by extrapolating linear regression of ASR vs. thickness to L = 0. Do this for each T, RH

Nafion® @ 30 oC

y = 181.891x + 0.056R2 = 0.970

y = 42.500x + 0.003R2 = 0.980

0.0

0.5

1.0

1.5

2.0

2.5

0.000 0.005 0.010 0.015Thickness, cm

Res

ista

nce,

Ω-c

m2

N111

N112

N1135

N11530% RH

90% RH

70% RH

50% RH

ASRcell

Unc

orre

cted

ASR

, Ω-c

m2

N = 3

Key Assumptions:1.Intrinsic through-plane conductivity is not a function of L2.Cell resistance is constant from build-to-build

R2 = 0.95 - 0.99 Similar ASRcell for 2

sets of PFSA membranesNafion® N1XX (4

thicknesses)Supported PFSA

membrane (3 thicknesses)

L = membrane thickness

Thickness, cm

21Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Cell resistance increases at low RH … dominated by interfacial resistance

1. Johnson & Liu, "Ionic Conductivity of Perflourosulfonic Acid Membranes as a Function of Temperature, Humidity and Equivalent Weight“ ECS PV 2002-5, 132 (2002)

Increasing Rcell with decreasing RH also reported by W.L. Gore & Associates 1

0.000

0.025

0.050

0.075

0.100

0.125

0.150

0.175

0.200

20% 30% 40% 50% 60% 70% 80% 90% 95%

Cel

l AS

R, o

hm c

m2

Relative Humidity

30C

80C

120C

120 oC

80 oC

30 oC

22Proprietary Data – Disclosure permitted only with prior written permission from Scribner Associates, Inc.

Ratio of cell to membrane resistance highlights importance of correcting for non-membrane ohmic contributions

At low RH, ratio is small relative to the membrane resistance Rcell : Rmembrane ~ 0.1 - 0.2

At high RH, the cell resistance can be significant relative to the membrane resistance, especially for thin membranes

0.0

0.1

0.2

0.3

0.4

0.5

0.6

20% 30% 40% 50% 60% 70% 80% 90% 95%

Relative Humidity

N111

N1135

N117

Relative Humidity

N135

N111

N117

R MembraneR Cell