Pore-size Dependence of Ion - PowerPoint Presentation

Slide1

Pore-size Dependence of Ion Diffusivity in Dye-sensitized Solar Cells

Yiqun MaSUPERVISOR: Dr. Gu Xu

1Slide2

Background and introductionDye-sensitized solar cellsMass transport in electrolyte

Problem: pore-size dependence of ion diffusivityExperimentalDevice fabrication and pore-size variationDC polarization measurementResults and discussionUnification of two opposite views

Unexpected surface diffusion

Significance of results

Conclusion

2

OutlineSlide3

Electrochemical cells utilizing dye molecules to harvest sunlightFirst published in Nature in 1991

7% overall power conversion efficiency was achieved, now has exceeded 12%New generation solar cell with possible low cost and high stability3

Introduction to Dye-sensitized

S

olar Cells

Oregan, B.; Gratzel

, M.,

Nature

1991,

353 (6346), 737-740Slide4

Monolayer Dye molecules for light absorption  High surface area required  mesoporous structure gives rise of

700 times of nominal surface areaWorking electrochemical Junction formed at the interface

4

Mesoporous

TiO

2 Thin Film

TiO

2

Dye

I

-

/I

3

-Slide5

5Device Physics of

Dye-sensitized Solar Cells

M

ass transport of ions

 Bottleneck of performance

FTOSlide6

6Three Possible

Mechanisms of Mass TransportKalaignan, G. P.; Kang, Y. S., J.

Photochem.

Photobiol

. C-Photochem. Rev.

2006, 7 (1), 17-22.

dominant

mechanism in DSSCs

In standard DSSCs, the mass transport rate is determined by the

diffusion of minority ions (I

3

-

) i.e. [I

3

-

] <<[I

-

]Slide7

Diffusion is pore-size independent when λ<0.1 (

λ = rmolecule/rpore)Based on the short

mean free path of inter-molecular collision in liquid

:

=

+

(

ε

: porosity;

τ

:tortuosity)

Tortuosity:

ratio of the length of the curve (

L

) to the distance between the ends of it (

C

)

7

Two Conflicting Views from Literature:

A) Pore-size

I

ndependent

D

iffusion

Karger

, J.; Ruthven, D. M., Diffusion in zeolites and other

microporous

solids. : Wiley: New York, 1992;

pp

350-365.

A

B

L

C

 Slide8

Frequently observed impeded diffusion in much larger pores (λ ≈ 0.01)

In this case ion diffusivity heavily depends on pore diameter8

Mitzithras, A.;

Coveney

, F. M.; Strange, J. H., J. Mol. Liq. 1992,

54 (4), 273-281.

40nm

Possibly due to the surface interaction or bonding

mechanisms

Decreases effective free pore volume

Two Conflicting Views from Literature:

B) Pore-size Dependent DiffusionSlide9

Remains controversial in dye-sensitized solar cellsYet critical

in estimation of the limiting current and design of efficient devicesBecause various fabrication processes lead to pore shrinkingDye loading

TiCl4

post-treatment

9

Debating

in

Dye-sensitized

S

olar CellsSlide10

Coating of Pt on FTO glass by heat treatment of chloroplanitic acid (H2

PtCl6) Deposition of TiO2 thin film by screen printing processSealing the cell with Surlyn film as spacer(25μ

m)

Injecting electrolyte (I

-/I3-

redox couple in acetonitrile) from the hole at the top

10

Experimental:

Device Fabrications

Injection hole

To focus on ion diffusion, a modified version of DSSC is

fabricatedSlide11

TiCl4 post-treatment is widely used in DSSC fabricationChemical bath which forms TiO2

on top of TiO2 mesoporous film by epitaxial growth – growing overlayer with the same structureReduce recombination rate and improve charge injection from dye molecules to

the CB of TiO2

Also reduce average pore size of TiO

2 film

11

Pore-size

Variation

by

TiCl

4 T

reatmentSlide12

12Pore-size

Variation by TiCl4 Treatment

Ito, S.; Murakami, T. N.; Comte, P.; Liska

, P.;

Gratzel, C.; Nazeeruddin, M. K.;

Gratzel, M., Thin Solid Films 2008,

516

(14), 4613-4619.

TiO

2

film on FTO/

Pt

glass

1

.

Immerse for 30

mins

2. Rinse with DI water

3. Anneal at

450

o

C

for 30

mins

Hot

Plate

0.1M

TiCl

4

aqueous solution at 70

o

C

TiCl

4

treated

TiO

2

film with smaller pores

TiCl

4

 + 2 H

2

O → TiO2 + 4 HClSlide13

13BET Characterization

Sample

Number of TiCl

4

treatments

Average pore diameter (nm)

Porosity ε

A

0

20.91±1.83

0.616±0.018

B

1

16.92±2.32

0.497±0.010

C

2

11.33±2.57

0.404±0.014

D

3

7.97±1.7

0.339±0.008

E

4

5.7±1.35

0.287±0.006Slide14

14BET CharacterizationSlide15

15Pore-size

Distribution

Curves follow more or less the normal distribution

Distribution

shape remains almost unchanged after treatments

Average pore diameter

decreases

Error bars of pore diameters are obtained from the FWHM values

Sample A, C and E underwent 0, 2 and 4 times of TiCl

4

treatments respectivelySlide16

Mass transport limited currentIn this case, diffusion limited currentIV curve will reach plateau at limiting current valueIn this case, the current will increase after the plateau

Charge injection from the TiO2 to electrolyte

16

DC P

olarization Measurement

I

V

I

lim

Ionic diffusion

Charge injection starts

V

T

The DC measurement was conducted in

DarkSlide17

First consider neat electrolyte between two electrodesAssuming

diffusion layer thickness = cell thickness, and

(since the current flow is independent of x)

General equation of diffusion limited current

F is the Faraday constant, c is the I

3

-

concentration and n is the stoichiometry constant which equals to 2 for I

-

/I

3

-

redox couple

17

Model

ConstructionSlide18

Continuity of current in the device: I = 2F

= 2FD

bulk

(

1)

The conservation of

I

3

-

ions:

c[

εt

+ (l – t)] =

ε

t+

(l – t)

(

2)

Combine (1) and (2) with boundary condition

c

0

=0:

= 4Fc

(

3)

18

Model

Construction

Kron

, G.; Rau, U.;

Durr

, M.;

Miteva

, T.;

Nelles

, G.; Yasuda, A.; Werner, J. H.,

Electrochem

. Solid State

Lett

.

2003,

6

(6), E11-E14.

t = 12

μ

m;

= 25

μm

 Slide19

19

DC Measurement Results

a) IV characteristic of control sample without TiO2 thin film;

b) Typical IV curves of samples A to E after 0 to 4 times of TiCl

4

treatments respectivelySlide20

Sample

I

lim

(mAcm

-2

)

D

TiO2

(10

-5

cm

2

s

-1

)

D

eff

(10

-5

cm

2s-1)

Tortuosity

(

)

A

35.25±1.25

0.747±0.038

1.22±0.09

1.05±0.09

B

24.80±0.60

0.513±0.016

1.03±0.05

1.24±0.06

C

21.10±0.45

0.437±0.012

1.08±0.07

1.18±0.08

D

16.67±0.35

0.343±0.009

1.01±0.05

1.26±0.06

E

10.33±0.50

0.207±0.011

0.721±0.055

1.78±0.13

Sample

I

lim

(mAcm

-2

)

D

TiO2

(10

-5

cm

2

s

-1

)

D

eff

(10

-5

cm

2

s

-1

)

A

35.25±1.25

0.747±0.038

1.22±0.09

1.05±0.09

B

24.80±0.60

0.513±0.016

1.03±0.05

1.24±0.06

C

21.10±0.45

0.437±0.012

1.08±0.07

1.18±0.08

D

16.67±0.35

0.343±0.009

1.01±0.05

1.26±0.06

E

10.33±0.50

0.207±0.011

0.721±0.055

1.78±0.13

20

DC

Measurement

R

esults

D

TiO2

:

ion

diffusivity in matrix

D

eff

: effective ion diffusivity normalized with porosity

: tortuosity calculated from

, expected to range from

1.2 to 1.8

*

 Slide21

21

Surprising Pore-size Dependence

A

B

C

D

E

D – E

:

P

ore-size

dependent

region,

D

eff

heavily depends on pore diameters

;

B

– D

: Pore-size independent region, almost forms a platform

;Transition:Critical point of transition is located at 5 – 7 nm;A – B: ? What is going on here?Slide22

22

Two Opposite Views Are Now Unified……Distinctive Regions of each diffusion mode

Pore-size dependent region

< 5 – 7 nm

Significant steric hindrance effect

of pore walls.Pore-size independent region

> 5 – 7 nm

Negligible collision

between

liquid molecules

and pore walls

Observed in DSSCs for the first time!

Pore-size dependent

Pore-size

independent

B

C

D

ESlide23

λ value at the transition ≈ 0.1 (550pm/5nm), which bears remarkable agreement

to the theoretical predictionThe range of pore-size independent region(>5-7nm) suggests fabrication processes of DSSCs will NOT cause transition of diffusion behaviorNot likely those processes will impede ion diffusivity significantly

23

……by the Critical Point of TransitionSlide24

24Significance of Our Results

Pore Size

Smaller

Large interfacial Area for efficient light harvesting

May impede mass transport rate

Larger

High mass transport limiting current

Not enough interfacial area

Our results

s

uggest the

minimum pore-size

without hindering the diffusion.

The balance between mass transport of electrolyte and interfacial area can be

optimizedSlide25

The tortuosity in A ≈ 1(unrealistic)

 Other diffusion mechanism is involvedSurface diffusionHopping mechanism of surface-adsorbed molecules between adsorption sites. Suppressed by the surface modification after

TiCl4

treatments

Act as a passivation process and decrease the number of available adsorption sites

25

Unexpected Rise from

B

to A

TiO

2

I

3

-

I

3

-

Surface diffusion

A

BSlide26

Both pore-size dependent and independent diffusion were observed under the same scheme by

altering the average pore-size of TiO2 matrix.The critical point of transition was located in the range of 5 – 7 nm. Thus standard fabrication processes will not cause transition of diffusion mode.Surface

diffusion mechanism was observed in

untreated TiO

2 and suppressed after the surface modification of TiCl4

post-treatment.

26

ConclusionSlide27

Dr. Gu XuDr. Tony Petric and Dr. Joey

KishDear group mates: Cindy Zhao, Lucy DengMr. Jim GarretDr. Hanjiang DongNSERC

27

AcknowledgementsSlide28

28Thanks for the attention!

Any questions?