1938-2

Workshop on Nanoscience for Solar Energy Conversion

27 - 29 October 2008

Quantum Dot Solar Cells: Semiconductor Nanocrystals As Light Harvesters

Prashant KAMAT University of Notre Dame Department of Chemistry 235 Radiation Laboratory Notre Dame IN 46556-0579 USA

Meeting the Energy Demand

14 TW Challenge

Quantum Dot Solar Cells. Semiconductor Nanocrystals as Light Harvesters Prashant V. Kamat
Dept Of Chemistry and Biochemistry and Radiation Laboratory Dept. of Chemical & Biomolecular Engineering University of Notre Dame, Notre Dame, Indiana 46556-0579
Support: US DOE (BES) http://www.nd.edu/~pkamat

After Oil
Beyond Fossil Fuels Farming Solar

Meeting the Energy Demand

L L I R D

D Y B A B

L L RI

www.solarsales.com.au

World Oil Production vs. Discovery Source: Dr. C.J. Campbell

Source:Wikipedia

Can we address the clean energy challenge with Nanotechnology?

Our Research Focus

Photoinduced electron transfer in light harvesting systems Donor- Acceptor Semiconductor-Sensitizer Semiconductor-Semiconductor Semiconductor-Metal

Molecular linker AcceptorPt

TiO
2

Donor

e
h e
Ag

e h

e h

CdSe TiO2

e h

C60

C60

C60

h
TiO 2

e h

ethanol products

Quantum Dot Solar Cells

Tunable band edge Offers the possibility to harvest light energy over a wide range of visible-ir light with selectivity

Hot carrier injection from higher excited state (minimizing energy loss during thermalization of excited state)

Multiple carrier generation solar cells. Utilization of high energy photon to multiple electron-hole pairs

m Aluminu

Glass
Redox or Hole Transport Layer Semiconductor Nanocrystals

M et al IT O

PE DO

m Aluminu

Metal Metal Junction Junction Solar Solar Cell Cell

Glass
Active Polymer/ Semiconductor Layer

IT O

CdSe TiO2 e OTE

h

Polymer Polymer Solar Solar Cell Cell

Pt

Electrolyte

Quantum Quantum Dot Dot Sensitized Sensitized Solar Solar Cell Cell

1. Semiconductor/metal Interface
ECB EF EVB h e

CB

et

+ +

ht

Red Ox

VB

Because of smaller dimensions charge separation and transport issues in nanostructure films need to be tackled

Metal/PbSeNC/Metal Sandwich Photovoltaic Cell

PbSe NC film, deposited via layer-by-layer dip coating Short-circuit photocurrent (>21 mA cm-2) by way of a Schottky junction EQE of 55-65% in the visible and up to 25% in the infrared region Power conversion efficiency of 2.1%.
Aluminum

Glass
Redox or Hole Transport Layer Semiconductor Nanocrystals

Luther et al Nano Lett., Vol 8, 2008, 3488

M et al IT O

2. Polymer Semiconductor Nanocrystal Hybrids

PE DO

m Aluminu

Glass
Active Polymer/ Semiconductor Layer

Adv. Mater, 2004, 16, 1009-1013

IT O

PE DO T

Aluminum

Hybrid Nanorod-Polymer Solar Cells

IT O

Glass
Active Polymer/ Semiconductor Layer

SCIENCE VOL 295, 2002 2425

3. Quantum Dot Sensitized Solar Cell (QDSSC)

e
h CdSe

TiO2

CB CB

e
h

e
TiO2

h e e e

e h e e h e e h e e h h

CdSe
VB VB

Red Ox
h

TiO2
e h

e h e h

CdSe

Photoelectrochemistry

Spectroscopy
detector

e
R O

h
pump probe

CB

CB

et

ht Red Ox

TiO2
VB

+ VB CdSe

GERISCHER H, LUBKE M A PARTICLE-SIZE EFFECT IN THE SENSITIZATION OF TIO2 ELECTRODES BY A CDS DEPOSIT JOURNAL OF ELECTROANALYTICAL CHEMISTRY 204 (1-2): 225-227 1986

detector

Photoexcitation of CdSe Quantum Dots

CB

pump

probe VB
1P(e)-1P3/2(h) 1S(e)-1S3/2(h)

Absorbance

0.5

-0.04

CdSe
440 nm 530 nm Time, ps

a
OD

OD

b
OD

440 nm 530 nm 3x 0.5 1.0 Time, ps

c d e
500 550

-0.08

a b c d e

0.2 ps 0.3 ps 0.4 ps 0.5 ps 1.1 ps

3x

0.5

1.0

400

450

600

650

Wavelength, nm

Charge Separation in TiO2/CdSe

Transient Bleaching Recovery of 3 nm CdSe Quantum Dots
Ex. 387 nm
CdSe-MPA
0.00

B
0.00

CdSe-MPA-TiO2

-0.04

-0.04 1 ps 35 ps 400 ps 1500 ps

-0.08

1 ps 35 ps 400 ps 1500 ps 450 500 550 600 650

-0.08

450

500

550

600

650

Wavelength, nm

Wavelength, nm

Modulation of the charge injection process by controlling the particle size?

e
CdSe 7.5 nm TiO2

e CdSe 2.4 nm

CdSe
e Slow Eg=1.9 eV h

TiO2

CdSe
e Fast

TiO2
e

Eg=2.4 eV h

Photoexcitation of CdSe Quantum Dots CdSe +h CdSe (ep +hp) CdSe (es + hs)
(A)
Absorbance Wavelength, nm 500 600

400 0.4 0.2 0.0

700

CdSe quantum dots of size 2.4 nm to 7,5 nm were excited with 387 nm laser pulse (130 fs) As the particle size decreases from 7.5 nm to 2.4 nm, the first (1S3/21Se) excitonic peak shifts from 645 nm (1.92 eV) to 509 nm (2.44 eV). Transient bleach corresponds to the first excitonic bleach

(B)
A

0.00

-0.02 d CdSe a 7.5 nm (Eg 1.92 eV) b 4.6 nm (Eg 1.99 eV) c 3.5 nm (Eg 2.18 eV) d 2.7 nm (Eg 2.35 eV) e 2.4 nm (Eg 2.44 eV) e

-0.04

c b 700

-0.06 400

500 600 Wavelength, nm

Robel, Kuno, Kamat, JACS 2007; 129, 4136

Electron transfer between CdSe and TiO2

CdSe (es) + TiO2 CdSe + TiO2(e)

0.8
size 7.5 nm (1S)=645 nm

a b

0.0 0.8
a size 4.6 nm (1S)=605 nm b

Normalized bleach

0.0 0.8
size 3.5 nm (1S)=570 nm a b a b a size 2.4 nm (1S)=509 nm

Analysis of Bleaching Recovery

0.0 0.8

A(t)=A(0)exp[-(t/)]
- where is the peak value of the characteristic lifetime

size 2.7 nm 0.0 (1S)=527 nm

0.8

0.0 0
Robel, Kuno, Kamat, JACS 2007; 129(14) pp 4136 - 4137

500

1000

1500

Time (ps)

Size Dependent Quenching Phenomenon

Diameter [nm] 7.5 4.6 3.5 2.7 2.3

Eg [eV] 1.92 1.99 2.18 2.35 2.44

CdSe [ps] 2332 7224 4420 6739 23119

CdSe

CdSe-TiO2 [ps]

CdSe-TiO2

ket [s-1]

0.697 0.474 0.417 0.457 0.51

2281 4961 1117 357 83

0.755 0.446 0.475 0.505 0.493

9.58106 6.3107 6.7108 2.65109 1.21010

A(t)=A(0)exp[-(t/)]

- where is the peak value of the characteristic lifetime

1/ 1/ = ket

Size Dependent Electron transfer between CdSe and TiO2

ket= 107 s-1 ket= 1.2x1010 s-1 e

ECB, V vs. NHE -0.9 -1.1

10
10

-1.3

2.4 nm 2.7 nm 3.5 nm

CdSe 7.5 nm

TiO2

CdSe 2.4 nm

10 ket, s
-1

10 10

4.6 nm
7

e
Eg=1.92 eV

e
Eg=2.4 eV

7.5 nm
0.4 0.6 -G ,eV 0.8

10

h
CdSe 7.5 nm

TiO2

h
CdSe 2.4 nm

Robel, Kuno, Kamat, JACS 2007; 129 pp 4136 - 4137

Linking Q-CdSe to TiO2 particles

e CdSe TiO2 h Electrode
HOOC SH HOOC SH HOOC SH OOC SH OOC SH OOC SH

b c

-0.75 nm 0 -0.75

2 m
150 nm

OTE

TiO2

Bifunctional linker molecule

Chemically Modified TiO2 film

OOC OOC SH OOC SH OOC SH OOC SOOC SS-

CdSe OTE/TiO2/CdSe Electrode

J. Am. Chem. Soc. 2006,128, 2385-2393

2 m

Photoelectrochemical behavior of Q-CdSe-TiO2 films

Potentiostat
WE RE CE

Voltage vs. SCE, V 0.2 Photocurrent, A 0 0.0 -0.2

TiO2

-0.4

-0.6

-0.8

-1.0

TiO2

R O

100

b
200
TiO2/CdSe

I-V characteristics of (a) OTE/TiO2 and (b) OTE/TiO2/MPA/CdSe films. Electrolyte 0.1 M Na2S. The filtered lights allowed excitation of TiO2 and CdSe films at wavelengths greater than 300 and 400 nm respectively

Modification of TiO2 Films with Different Size CdSe Particles

2.3nm 2.6nm

3.0nm 3.7nm
3.7 nm 3.0 nm 2.6 nm 2.3 nm

2.3nm 3.0nm 3.7nm 2.6nm

0.6

0.9

Absorbance

0.4

Absorbance

3.7 nm 3.0 nm 2.6 nm 2.3 nm

0.6

0.2

0.3

0.0 300 400 500 600 700

0.0 450 500 550 600 650 700

Wavelength, nm

Wavelength, nm

Tuning the Photoresponse of Quantum Dot Solar Cells

e IPCE or Ext. Quantum Eff.
R O

= (1240/) x (Isc/Iinc) x 100

E
50 40

(A)

3.7 nm 3.0 nm 2.6 nm 2.3 nm

e
CB

IPCE (%)

30 20 10 0 350 400 450 500 550 600 650 700

h
RO

VB

CdSe

TiO2
Wavelength (nm)
J. Am. Chem. Soc., 130 (12), 4007 -4015, 2008

Photocurrent Response
Efficiency of Charge Injection vs. Light absorption

Current density, mA/cm

(A)

R O

3.7 nm 3.0 nm 2.6 nm 2.3 nm

0 0 50 100 150

Solar Flux

Time, sec

Can we employ the nanowire/nanorod architecture to improve the performance of quantum dot solar cells?
CdSe CdSe
e e e e

TiO2

OTE/TiO2 nanoparticles (a)

Ti/TiO2 nanotubes (b)

Recent advances
Nanowire dye-sensitized solar cells
LAW, GREENE, JOHNSON, SAYKALLY, YANG Nature Materials 4 , 455, 2005

Fast Electron Transport in Metal Organic Vapor Deposition Grown Dye-sensitized ZnO Nanorod Solar Cells
Galoppini, Rochford, Chen, Saraf, Lu, Hagfeldt, and Boschloo J. Phys. Chem. B; 2006; 110 16159

Electron transport in solar cells with ZnO-nanorod electrodes was about 2 orders of magnitude faster (30s) than ZnO-colloid electrodes

Mor, G. K. et al Use of highly-ordered TiO2 nanotube arrays in dye-sensitized solar cells. Nano Lett., 2006. 6, 215-218.

Leschkies, K. S et al Photosensitization of ZnO nanowires with CdSe quantum dots for photovoltaic devices. Nano Lett., 2007. 7, 1793-1798.

Martinson, A. B. F. et al., ZnO nanotube based dye-sensitized solar cells ZnO nanotube based dye-sensitized solar cells. Nano Lett., 2007. 7, 2183-2187.

200nm C D

5m

2m

500nm

(B) (A)

2.3nm 3.0nm 3.7nm 2.6nm

2.3nm 2.6nm

3.0nm 3.7nm

(C)
2.3nm 3.0nm 3.7nm 2.6nm

(B) 3.7 nm CdSe

10000

Intensity (a.u.)

CdSe/glass

Photoemission of CdSe/glass and CdSe/TNP

2.6 nm diameter a. CdSe/glass b. CdSe/TiO2 3.7 nm diameter c. CdSe/glass d. CdSe/TiO2

1000 100 10 1 0 5
prompt
CdSe/TiO2(NP) CdSe/TiO2(NT)

1.2

10

15

20

25

Intensity (a.u.)

Time, ns

0.8

<t>, ns
CdSe/ glass
650 700

ket, sec-1

0.4 b 0.0 500

3

2.6nm 3.7nm 2.6nm 3.7nm 2.6nm 3.7nm

4.1 7.9 0.4 1.3 0.4 1.5

2.5E+09 6.3E+08 2.2E+09 5.5E+08

d
3

550

600

Wavelength, nm

CdSe/ tnp CdSe/ tnt

Emission spectra of CdSe QDs (a, c) on glass and (b, d) chemically bound to TiO2 nanoparticle films at 2 different sizes of QDs (2.7 and 3.7 nm). Excitation was at 480 nm.

Quantum Dot Solar Cells Particle versus Tube Architecture

50 40

(A) TiO2(NP)

IPCE (%)

IPCE (%)

30 20 10 0 350 400 450

d c

CdSe Diameter 3.7 nm a. 3.0 nm b. 2.6 nm c. 2.3 nm d.

50 40 30

(B) TiO2 (NT)

d c b

CdSe Diameter 3.7 nm a. 3.0 nm b. 2.6 nm c. 2.3 nm d.

b a

20
a

10 0

500

550

600

650

700

350

400

450

500

550

600

650

700

Wavelength (nm)
200nm

Wavelength (nm)

500nm

ISC mA/cm2
CdSe-TNP CdSe-TNT 1.64 1.95

VOC V
0.591 0.582

Pmax mW/cm2
0.25 0.29

FF

0.26 0.26

J. Am. Chem. Soc., 130 (12), 4007 -4015, 2008

Power conversion efficiency ~1%

Depositing CdS quantum dots on TiO2 nanotubes

Unsonicated TiO2 NT

0 dip

5 dip

10 dip

20 dip

100kX

500kX

Scale bars: Top 500nm, bottom 50nm

(TiO2)

Cd2+

(TiO2)Cd2+

Wash

S2-

(TiO2)CdS

Wash

1 dip cycle

Photocurrent Response of TiO2 (nanotube)CdS Films

IPCE: CdS on Single TiO2 electrode
60

50 40 IP C E (% ) Tubes 30 5 dip 10 dip 20 dip 20 10

R O

0 350

400

450 Wavelngth (nm)

500

550

600

Reflectence Absorbtion Spectra: CdS/P25/OTE

1 0.9 0.8 0.7 Absorbance 0.6 0.5 0.4 0.3 0.2 0.1 0 350 TiO2 5dip 10dip 15dip 20dip

400

450

500

550

600

Wavelength (nm)

Carbon nanostructures as conduits to transport charge carriers

Advantages
High surface area Good electronic conductivity, excellent chemical and electrochemical stability Good mechanical strength

Goal
Effective utilization of carbon nanostructures for improving the performance of energy conversion devices - To develop electrode assembly with CNT supports - Improve the performance of light harvesting assemblies - Facilitate charge collection and transport in nanostructured assemblies

Pt

..towards achieving ordered assemblies on electrode surface

SWCNT- TiO2 composite films

Mesoscopic TiO2 films are extensively used in Dye-Sensitized Solar Cells A carbon nanotube support architecture can disperse the TiO2 particles and facilitate charge collection and charge transport within the film. The first step is to design the SWCNTTiO2 network and test the feasibility of the composite system in solar cells
h

e e
CB

et

+ +

ht

VB

Kongkanand, A.; Domnguez, R.M.; Kamat, P.V., Single Wall Carbon Nanotube Scaffolds for Photoelectrochemical Solar Cells. Capture and Transport of Photogenerated Electrons. Nano Lett., 2007. 7, 676-680. Vietmeyer, F.; Seger, B.; Kamat, P.V., Anchoring ZnO Particles on Functionalized Single Wall Carbon Nanotubes. Excited State Interactions and Charge Collection. Adv. Mater., 2007, 19: 2935-2940

Electrophoretic Deposition of SWCNT on Electrode Surfaces

Sonication for 1 min SWCNT+TOAB in THF

After Electrodeposition on OTE OTE/SWCNT

SWCNT/TOAB in THF

0.8
B

0.6 Abs 0.4

OTE/SnO2/SWCNT

b
OTE/SWCNT

200 nm

0.2 0.0 400

OTE/SnO2

600 800 1000 Wavelength, nm

J. Am. Chem. Soc., 2004. 126 10757-10762.

A
Carbon Fiber Paper (CFE)
CFE

50 m

5 m

B
TiO2 Deposition On CFE
CFE/TiO2

50 m

5 m

C
SWCNT Deposition On CFE
CFE/SWCNT 50 m

1 m

D
SWCNT-TiO2 on CFE
CFE/SWCNT/TiO2 50 m

1 m

Photocurrent Generation
CFE/TiO2 versus CFE/SWCNT TiO2
Photocurrent, A/cm
60

a) SWCNT/TiO2 b) TiO2

40

b
20

0 0 20 40 60

Time, sec

UV light

15

a b c

Higher IPCE (increase of factor ~2) was observed for mesoscopic CFE/SWCNT-TiO2 films The results are indicative of better charge collection and transport provided by the SWCNT -Network

IPCE, %

10

a) SWCNT/TiO2 at 0V b) SWCNT/TiO2 at SC c) TiO2 at 0V d) TiO2 at SC

0 350 400 450 500

Wavelength, nm

Nano Lett., 2007. 7, 676-680

Dependence of TiO2/SWCNT Ratio on the Photocurrent Generation

60

Photocurrent (A/cm )

50 40 30 20 10 0

1 m
a) SWCNT/TiO2 b) TiO2

3
2

TiO2 loadings (mg/cm )

Increasing the TiO2 concentration results in enhanced photocurrent as they are dispersed on SWCNT network. At concentrations greater than 2 mg/cm2 the beneficial effect of SWCNT disappears. Under these conditions. TiO2 particles aggregate and the charge recombination dominates Nano Lett., 2007. 7, 676-680

Where do we go from here?

Capping CdSe with an Electron Acceptor Shell

C60 CdSe CdSe C60 1 m
1.2 m

Electrophoretic deposition of Cluster films

0
5 (a) 4 (b) (c) nC60

0 1 m

CdSe CdSe/nC60

IPCE (%)

1 a 0 350 400 450 500 550 600 650 700

Wavelength (nm)

J. Am. Chem. Soc., 2008, 130, 88908891

Organized light harvesting assembly using carbon nanostructures

h

e h

1 m
15 nm

e h

e h

0 1 m

Graphene-Semiconductor Nanocomposites

ACS Nano, 2008, 2, 1487-1491

Graphene Oxide

h h TiO TiO GO 2
2

TiO2-GR

0.6 m

Reduced Graphene

Summary

Unique properties of quantum dots offer new opportunities to develop low-cost and high efficiency solar cells 1-D architectures are useful for designing next generation solar cells. Opportunities exist for carbon nanostructures to facilitate capture and transport of electrons in nanostructure semiconductor based solar cells.

Kamat, P. V. Meeting the Clean Energy Demand: Nanostructure Architectures for Solar Energy Conversion (Review) J. Phys. Chem. C, 2007. 111 2834 - 2860. Quantum Dot Solar Cells. Semiconductor Nanocrystals as Light Harvesters (Centennial Feature) J. Phys. Chem. C 2008, 112, in press

Researchers/Collaborators
Graduate students

Brian Seger (Chem. Eng.) David Baker (Chem. Eng.) Kevin Tvrdy (Chemistry) Clifton Harris (Chemistry) Matt Baker (Physics) Ian Lightcap (Chemistry) Philix Vietmeyer (Chemistry) Yanghai Yu (Chem. Eng.) Istvan Robel (Physics)

Post-Docs/Visiting Scientists

Jin Ho Bang

Pat Brown Chris Rodriguez David Riehm Rachel Staran

Undergraduate students

Collaborators Dr. K. G. Thomas (India) Prof. Fukuzumi (Osaka U.) Prof Ken Kuno (UND) Prof. K. Vinodgopal (IUN)

What will the future hold?

Over the last twenty years, the per-kWh price of photovoltaics has dropped from about $500 to nearly $5; think of what the next twenty years will bring.

http://www.theleveredge.com/images/isw_cartoon.gif