Mesoscopic Injection Solar Cells for ElectricityGeneration from Sunlight

Michael GraetzelSwiss Federal Institute of Technology Lausanne

michael.graetzel@epfl.ch

International Materials ForumBayreuth Germany August 1-2, 2005

• Nanocrystalline Films: Dr. L. Cevey, Pascal Comte, Francine Duriaux- Arendse, Raphael Charvet, Dr.Carole Graetzel, Peter Chen

• Dye Research: Dr. M. K. Nazeeruddin, Dr. S. M. Zakeeruddin, Dr.Cédric Klein, Dr. Nick Evans Dr. Peter Pechy, Anthony Burke

• PV cells : Dr. Peng Wang. Dr. Lukas Schmidt-Mende, Dr.P. Liska, Dr. Seigo Ito, Takeru Bessho, Dr. Robin

Humphry-Baker, Nathalie Rossier, Dr. Henry Snaith• Electrochemistry: Dr. Qing Wang, Dr. Davide Dicenso, Ilkay Cesar, Shipan Zhang• Electron transfer: Dr. Jacques-E.Moser, Bernard Wenger, Dr.

K.Kalyanasundaram• Modeling, analysis Dr. Guido Rothenberger, Dr Pierre Infelta, Dr.

François Rotzinger• DFT calculations: Filippo De Angelis, Simona Fantacci (Perugia), Annabella

Selloni (Princeton).We are grateful for funding from:

The Swiss Top Nano-21program,The Swiss Energy OfficeSwiss National Science FoundationUS Air Force (European Office of Aerospace Research and Development)European Joule Projects (NANOMAX, MOLYCELL) , European Cost D14Industrial Partners,

Humanity’s Top Ten Problemsfor next 50 years

1. ENERGY2. WATER3. FOOD4. ENVIRONMENT5. POVERTY6. TERRORISM & WAR7. DISEASE8. EDUCATION9. DEMOCRACY10. POPULATION

2003 6.3 Billion People2050 8-10 Billion People

Source Richard Smalley Energy & Nanotechnology ConferenceRice University, Houston May 3, 2003

THE SOLAR CHALLENGE

● With a projected global population of 12 billion by 2050 coupled with moderate economic growth, the total global energy consumption is estimated to be ~28 TW. Current global use is ~11 TW.

● To cap CO2 at 550 ppm (twice the pre-industrial level),most of this additional energy needs to come from carbon-free sources.

● Solar energy is the largest non-carbon-based energysource (100,000 TW).

● However, it has to be converted at reasonably low cost.

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.00

10

20

30

40

50

60

70

80

0

10

20

30

40

50

60

70

80

Sola

r P

hoto

n F

lux (

mA

/cm

2

.eV

)

Energy (eV)

6000K BB integrated current

AM1.5G integrated current

6000K Blackbody Spectrum

100 mW/cm2

!(E) = AM1.5G Solar Spectrum

100 mW/cm2

Inte

gra

ted p

hoto

n flu

x (

mA

/cm

2

)

Solar Spectrum and Available PhotocurrentTHE SOLAR RESOURCE

Photovoltaic CellsI. 1st Generation

• Single crystal Si• Poly-grain Si

II. 2nd Generation (Low Cost--Mainly Thin Films)• Amorphous Si• Thin film Si• CuInSe2• CdTe• Dye-sensitized Photochemical Cell• Organic PV (molecular and polymeric)

III. 3rd Generation (ntheor>31%(the Queisser-Shockley limit))• (High efficiency multi-gap tandem cells (already here))• Hot electron converters• Carrier Multiplication cells• Mid-band PV• Quantum Dot Solar Cells• Other approaches

Photovoltaic market growth projection until 2030

300 GW/Year@ 2030

30 % p.a.

25 % p.a.

Courtesy Dr. Winfried Hoffman, CEO, RWE, SCHOTT Solar GmbH

PRESENT PV TECHNOLOGY

(Dominated by semiconductor p-njunctions)

e-

usable photo-voltage (qV)

Energy

e-

n-typep-type

1 e- - h+ pair/photon; full hot carrier relaxation

ηmax = 32%

(in radiative limit)heat loss

heat loss

hν

Conventional Single Conventional Single HomojunctionHomojunction PV Cell PV Cell

h+

Production Forecast of Solar Modules Using DifferentTechnologies

2010 (Forecast)

Jp 1.200EU 1.000US 500SOA 500ROW 500Σ 3.700

MW GW30%p.a. 25%p.a.

Courtesy Dr. Winfried Hoffman, CEO, RWE, SCHOTT Solar GmbH

Market Size in 2030 for the four market segments

Rural Electrification

60 GWp p.a.On-Grid

150 GWp p.a.

Consumer

20 GWp p.a.

Off-Grid Industrial

70 GWp p.a.

Total 300 GWp⇒ 200 bio €

Modul Turnover in 2030

Courtesy Dr. Winfried Hoffman, CEO, RWE, SCHOTT Solar GmbH

Emerging and new applications call for:

• colour• flexibility• light weight• easy of integration• many more

... further development and new technologies inorder to meet optimally the customer demandsand needs

Courtesy Dr. Winfried Hoffman, CEO, RWE, SCHOTT Solar GmbH

Band Diagram

Mesoscopic Injection Solar Cells

Scheme

M. Grätzel, Nature 2001, 414, 338−344

I3-

3I-

Sensitized Sensitized mesoscopic heterojuntionsmesoscopic heterojuntions

TiO2

hhννTi-

Ti-

3I-/I3-

N

N

N

N N

Ru

N

OHO

O

HO

C

C

O

HO

OHO

S

S

e-

e-

hhνν Ion Diffusion

DyeEVB

ECB

S+/S

S+/S*

M. Grätzel, Nature 2001, 414, 338−344.

B. O’Regan, M. Grätzel, Nature 1991, 353, 737−740

Device Concept: Device Concept: ““solid-statesolid-state””

hhννHole Hopping

η = 4 % @ 1 Sun

spiro-MeOTAD

U. Bach, D. Lupo, P. Comte, J.-E. Moser, F. Weissörtel, J. Salbeck, H.Spreitzer and M. Grätzel, NATURE 395, 583-585 (1998)

R.D. Schaller and V.I. Klimov, Phys. Rev. Letts, 92,186601 (May), 2004 (PbSe QDs)

1 percolation of electrons and holes2 selective contacts3 no electrical shunts (pinholes)4 stoichiometric TiO2 5 p-type CuInS2

6 buffer layer7 intimate contact8 complete filling of pores

3D solar cells

M. Nanu, J. Schoonman, and A. Goossens, Advanced Materials 16 (2004) 453M. Nanu, J. Schoonman, and A. Goossens, Adv. Func. Mat. 15 (2005) 95

Bulk organic heterojunction solar cell

Glass

PEDOT:PSS

Active layer

LiF

AluminumSMU

+-

ITO

Illumination

conductingpolymer

Alpolymer / fullereneblend

TCO

-+

O

OMe

PCBM blended with:

JSC = 9.3 mA/cm2

VOC = 0.56 VFF = 60%η = 3.5%

OC1C10-PPV

O

O

(

)n

S

S

n

JSC = 5.3 mA/cm2

VOC = 0.82 VFF = 61%η = 2.5%

(under AM1.5 illumination)

P3HT

0.6

0.4

0.2

Quantu

m E

ffic

iency

800700600500400

Wavelength (nm)

0.3 s integration, 7 counts 1.0 s integration, 9 counts

From G. Rumbles NREL

Outline

• Mesoporous junctions, interfacial and cross-surface charge transfer

• Photoinduced charge separation• Photogalvanic generation of electricity from

sunlight

Mesoscopic semiconductor filmsexhibit extraordinary properties

• surface amplification ca 100 times for eachmicron film thickness

• Interpenetrating network electronic junctionhaving huge contact area

• ease of electron percolation through the particlenetwork

• very rapid lithium insertion and release• high photocatalytic activity• high sensitivity for detecting ambients• efficient photovoltaic energy conversion

Rapid electron percolation through nanocrystalsRapid electron percolation through nanocrystals

+ ++

+

+

+

+++

+-- -

-- -

--

--

+

-

Charge of electrons compensated by inert positive ions in electrolyte

µµee

G. Rothenberger, M Grätzel and D Fitzmaurice, J.Phys Chem.1992,96,5983

No space charge limitation of current !

Kavan M. Grätzel Electrochemical and Solid State Letters 5 (2): A39-42 (2002)

Cross surface electron and holetransfer in self-assembled molecular

charge transport layers

Molecular wiring of insulating Nanocrystals

Ambipolar lateral charge percolationin self-assembled monolayers on

nanocrystalline insulator films

Z907 sensitizer

P.Wang, S.M. Zakeeruddin, R. Humphry-Baker, J.-E. Moser, M. GrätzelAdv. Materials, 15, No. 24, 2101-2104 (2003)

HOMO LUMO

A. Hagfeldt, M. Grätzel, Acc. Chem. Res. 2000, 33, 2679−27

Cyclic Voltammogram of Z 907 on Aluminium Oxide Film

Electrolyte; EMITFSI in Acetonitrile, Scan rate= 0.1V Sec-1

Cross surface hole percolation through a self-assembled Z-907 monolayeradsorbed on mesoscopic anatase TiO2.

QING WANG, ROBIN HUMPHRY BAKER AND MICHAEL GRAETZEL to be submitted

Cyclic voltammogram of the N621 complex anchored onto aTiO2 electrode (black line), exposed to the solutions of HgCl2

(red dashed dotted line) and Pb(ClO4)2 (blue dashed line)

600x10-6

400

200

0

-200

-400

i / A

1.41.21.00.80.60.40.2

E / V vs. Ag QRE

Molecular structure of the N719-HgCl2 complex showing the asymmetric unit which contains one mercury atom, and two Cl– anions

Outline

• Mesoscopic junctions, interfacial and cross surfacecharge transfer

• Photoinduced charge separation• Photogalvanic generation of electricity from

sunlight

Silicon Photovoltaic Cells Dye Solar Cells

Charge separation by electricfield within a p- and n-dopedsemiconductor material (Si, II-VI, a-Si: H)

Charge separation by kineticcompetition like inphotosynthesis

EF

p-Si (B)[CdTe]

D/D+

e--R

n-Si (P)[CdS]

hν hν

D*/D+TiO2

Electrolyte

Dye

cb

The two dilemmas of light harvesting bysurface immobilized molecular absorbers

1. A monolayer of dye on a flat surface absorbs at mosta few percent of light because it occupies an area that ismuch larger than its optical cross section

2. Compact semiconductor films need to be n-doped toconduct electrons. Energy transfer quenching of theexcited sensitizer by the electrons in the semiconductorleads to conversion of light to heat reducingphotovoltaic conversion efficiency.

Anatase crystals

• Undoped crystal,(001) surface

• Doped crystal,(101) surface

A. Vittadini, A. Selloni, F. Rotzinger and M. Grätzel Phys. Rev. Lett. 81, 2954 (1998)

Incident photon to electron conversion efficiency (IPCE)of a dye-sensitized TiO2 (101) single crystal PEC solar cell

IPCEmax = 0.13 %

IPCEmax = 88 %

Incident photon to current conversion efficiency of a dye-sensitized solar cell based on a mesoscopic TiO2electrode

Dye sensitized nanocrystalsshow quantitative conversionof the photons into electriccurrent

Competition ⇒

Electron diffusion length

Dynamic CompetitionDynamic Competition

electron transport

loss mechanism:interfacial

recombination

!

Ln= "Dn τn

electron injection

dye regeneration

electron transport

interfacialrecombination

time [s]

τn: electron lifetimeDn: electron diffusion coefficient

Photo Induced Heterogeneous ElectronTransfer Cycle

N

N

Ru

C

C

O

- O

- O

O

OXIDE

Tis4+

–

–

– MLCT EXCITATION

forward reaction

Backward transfer

–

–

–

Ru (II/III)

dxy, dxz, dyz

!LIGAND *ORBITAL

h ! (" 1.7 eV)

kf

Ti 4+/3+

spatial contraction of d orbitals upon

oxidation from Ru(II) to Ru(III)

Energy

kb

Finite length transmission model (Bisquert)

Chemical Capacitance

Transport

Recombination

Q Wang, J. Moser and M. Graetzel J.Phys. Chem B in press

Electron Transport: Diffusion and Electron Lifetime

e-

e-I3-

Electrons should travel to the SnO2before charge recombination occurs

Diffusion length should exceed thethickness of the mesoscopic TiO2 film

!DL =

SnO2:F TiO2/Electrolyte

10~20 µm

Charge recombination2 e- + I3- = 3I-

Increasing the injection and lowering the recombination rates

is critical for maximizing the open circuit voltage of the cell !

Voc = (nRT/F)ln(KΦ/(k1[S+] +k2[D+])

KΦ : charge carrier photo-generation rate

k1, k2: recombination rate constants

n: ideality factor of the junction

-15

-10

-5

0

Cu

rren

t D

en

sity

[mA

/cm

2]

0.80.60.40.20.0Potential [V]

•

100% AM1.5Efficiency 11.04%

•65% AM1.5

•9.5% AM1.5

Efficiency 11.18%

Efficiency 10.87%

STABILITYRequirements for outdoor use according tointernational PV standards applied to single crystalsilicon but so far not to thin film PV cells

UV plus heat (55-60 C): 1000 hours

Accelerated thermal test at 85 C: 1000 h

Humidity test and temperature cycling (sealingissues)

Self-assembly of stable and welldefined monomolecular layers ofsensitizer at the interface provideslong term photovoltaic stability andhigh conversion efficiency

Interface Engineering Interface Engineering in Dye-Sensitised Solar Cellsin Dye-Sensitised Solar Cells

ROBUSTElectrolyte

PMII: 0.8 M

I2: 0.15 M

NMBI: 0.5 M

0.1 M GSCN

MPN solvent

Efficiency: > 8.0%

Photoanode: 8+5K-19

Decylphosphonate

80 oC evolution of device parameters in the dark

Wang, P.; Klein, C.; Humphry-Baker, R.; Zakeeruddin, S. M.; Grätzel, M. Appl. Phys. Lett. 2005, 86. 123508.

60 oC evolution of device parameters under one sun soaking

SOLVENT-FREE SYSTEMS

SOLID (POLYMER)ELECTROLYTES,SOLIDIFIED IONIC LIQUIDSHOLE CONDUCTORS

Ionic solid electrolytes

P. Wang, S.M. Zakeeruddin, P. Comte, I. Exnar, and M. GrätzelPeng Wang et al J. Am. Chem. Soc 2003, 125, 1166-1167

Consists of only IonsLiquid under wide temp. range ex. -10℃ to 400℃non volatileChemically stable and non combustibleHigh electronic conductivity

NNCH2CH3H3C

+

O

F3C-

SS N CF3

OO

O

1-Ethyl-3-methylimidazolium - Bis(trifluoromethylsulfonyl)Amide

EMIm-TFSA

Features of Ionic LiquidsION-GEL Electolyte (NEDO)

Nano composite Ion Gel

EMIm-TFSA + TiO2 (P25, 10 wt% )

Centrifugal separation2000 G (6400 rpm) x 1hr

Ionic liquid

TiO2nano

composite gel

I2 content

References(1) T.Fukushima, A.Kosaka, Y.Ishimura, T.Yamamoto, T.Takigawa, N.Ishii and

T.Aida, Science, 27(2003)2072.(2) P.Wang, S.M.Zakeeruddin, P.Comte, I.Exnar and M.Graetzel, J.Am.Chem.Soc.,

125(2003)1166

Fig. I-V characteristics of nano composite ion gel cell and bare ionic liquid cell

4.80.5770612.0MW-CNT

5.80.6471912.5SiO2

4.70.6066111.8withoutparticles

5.10.6068512.5ITO5.00.6167912.1SnO2

5.70.6569612.5TiO2

η(%)

FFVoc

(mV)Jsc

(mA/cm2)nano

particles

Table PV performance of nanocomposite

ion gel cells

TiO2 (P25); anatase, 28nm

-5

0

5

10

15

0 200 400 600 800

Voltage (mV)

Cu

rre

nt d

en

sity (

mA

/cm

2)

Ionic liquidEMIm-TFSA η = 4.7 %

TiO2 Nanocomposite gel

η = 5.7 %

Influence of Nano Particles

Courtesy of Dr. Nobuo Tanabe Fujikura Ltd

Dye /TiO2 layer

Nanocompositelayer

Photoelectrode Counterelectrode

Grotthus-like exchange mechanism

Nanoparticle

Iodide ion (I-)

Tri-iodide ion (I3-)

Charge transfer path

Imidazolium cation

e-

I3- I-

Charge transportationin nano composite gel

Chem.Commun. 2005,363-365

Viscosity: 900 cp at 22 oC

Viscosity: 18 cp at 22 oC

Photovoltaic performance

7.7%

Wang, P.; Klein, C.; Humphry-Baker, R.; Zakeeruddin, S.M. andGrätzel.M. J. Am. Chem Soc. 2005, 127, 808.

Evolution of device parameters using quasi-solid ionic liquidgel electrolyte under one Sun light soaking at 60 oC

NH

HN

HN

O

O

C14H29

PMII/EMINCS: 65:35 (volume)

I2: 0.2 M

NMBI: 0.5 M

GuNCS: 0.1 M

Gelator: 2 wt %

Gelator

0 10 20 30 40 504

6

8

10

12

14

Time (d)

! (

%)

J s c

mA

cm

- 2

0.4

0.5

0.6

0.7

0.8

ffV

o

c (V)

Dye K19+DPA

N

N

N

N

Ru

NCS

NCS

OHO

OH

O

O

O

O

O O

O

O

O

Ion coordinating sensitizers

Immobilization of Li Ions

K51

Z907Z907

K51

no Li × 10

Li hugely increases Jdensity

No longer space-charge limited current.

K51 immobilizes ionson dye backbone.

N

N

N

N

Ru

NCS

NCS

ONaO

OH

O

O

O

O

O

O

O

O

O

Li+

Li+

0

5

10

15

20

25

30

35

40

0 0.5 1 1.5

Cur

rent

Den

sity

(mA

cm-2

)

Applied Bias (V)

56789

10

20

30

40

0.1 1 10 100

Curr

ent D

ensi

ty (m

Acm

-2)

Time (S)

Temperature Dependence

Increased charge mobility with temperature ↑ current and fill factor

Increased charge recombination ↓ voltage and current

For K51 ↑ mobility wins. For Z907 ↑ charge recombination wins

-1.2

-1

-0.8

-0.6

-0.4

-0.2

00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

K51 device 291278274264250240230220210200

Cur

rent

Den

sity

(mA

cm-2

)

Applied Bias (V)0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Z907 device 293283276263253235225215204200

Applied Bias (V)

Transient Voltage Decay’s

0.1

1

0 0.2 0.4 0.6 0.8 1 1.2

Nor

mal

ized

ΔV

Time(mS)

Lifetime(ms)0.45K510.32Z9070.03K51 no Li0.01No Dye/Li

Advantages vs. Silicon Cells• Low cost and ease of production• Performance increases with temperature narrowing the

efficiency gap• Bifacial configuration - advantage for diffuse light and

albedo• Efficiency less sensitive to angle of incidence• Transparency for power windows• Color can be varied by selection of the dye, invisible PV-cells

based on near-IR sensitizers are feasable• Low energy content (for silicon this is 5 GJ/m2 !), payback

time is only a few months as compared to years for silicon.• Outperforms amorphous Si

UltimateThermodynamic

limit at 1 sun

min BOS

Shockley-Queisser limit

PV Power Costs as Functionof Module Efficiency and Cost From Martin Green

For PV to provide the full level of C-free energy required for electricity andfuel—solar power cost needs to be ~5 cents/kWh ($1.00 Wp)

Courtesy of Greatcell Solar

Various colours in a series-connected dye solar cell module

Courtesy Dr. Winfried Hoffman, CEO, RWE, SCHOTT Solar GmbH

Hitachi’s new dye sensitized cell achieves 9.3 percent efficiency

DSCmade by

AISIN -SEIKI http://www.toyota.co.jp/jp/news/04/Dec/nt04_1204.html

The Toyota Dream House

Future Generation PV TechnologiesFuture Generation PV Technologies• Opportunities for innovation– ultra low

cost, ultra high efficiencies, newmaterials, new chemistry,new physics- Photoelectrochemistry- Excitonic cells (dye cells,semiconducting polymers, molecularsemiconductors)- Quantum dot cells; hot-carrier andimpurity band cells; photon tailoring

Ideal StructureIdeal Structure

GlassMo

NanoparticulateCIGSprecursormaterials

From T. Surek

Integration of Photosynthetic Protein Molecular Complexes in Solid-State Electronic DevicesRupa Das, Patrick J. Kiley, Michael Segal, Julie Norville, A. Amy Yu, Leyu Wang, Scott A. Trammell, L. Evan Reddick, Rajay Kumar, Francesco Stellacci, Nikolai Lebedev, Joel Schnur, Barry D. Bruce, Shuguang Zhang, and Marc BaldoNanoletters 2004, vol 4, pp 1079 - 1083;

Can laptops run on spinach ?

Spinach photosynthetic power can create electricity.