Applied nanotechnology
- the dye-sensitised solar cell
Augustin McEvoy
Lab. for Photonics and Interfaces,
Faculty of Basic Sciences
ÉCOLE POLYTECHNIQUE
FÉDÉRALE DE LAUSANNE
Aarhus, Denmark
August 2007
What’s nanotechnology?
Definition
Encyclopædia Britannicana: nanotechnology
“Manipulation of atoms, molecules, and materials to
form structures on the scale of nanometres
(billionths of a metre)”.
To scale:
1 nanometer = 10-9 meter (nm)
1 micrometer = 10-6 meter (µm)
Average human hair = 50µm
DNA helix = 2nm.
Atomic separation in a crystal = 0.2 nm
20th U.S. edition, Gray's Anatomy of the Human Body, (1918)
Nanotechnology isn’t new!
Nanotechnology can be seen as an extension of existing
sciences into the nanoscale, or as a recasting of existing
(Wikipedia)
sciences using a newer, more modern term.
The word ”Nanotechnology" is due to Norio Taniguchi
(Tokyo Science University)
(N. Taniguchi, "On the Basic Concept of 'Nano-Technology'," Proc. Intl. Conf. Prod.
Eng., Tokyo, Part II, Japan Society of Precision Engineering, 1974.)
But recent research shows that “Damascene” sword
blades - over 1000 years old - may owe their properties
to a combination of cementite (iron carbide) and carbon
nanotubes. ReiboldM., et al. (TUDresden), Nature, 444 (2006) 286.
Impetus for nanotechnology
Nano-scale imaging and analysis:
X-ray diffraction - crystal structures - von Laue, 1912.
ESCA - surface chemical analysis - K.Siegbahn, 1954.
Actual nanoscale imaging had to await the atomic force microscope (AFM) and
scanning tunneling microscope - Binnig,Rohrer - 1981.
Nano-scale fabrication
Electron beam lithography, using patterning to control structure - computer
chips are manufactured with device sizes of 45 nm.
Molecular beam epitaxy (MBE) and Atomic Layer Deposition (ALD) provide
artifacts on a subnano scale.
Chemical synthesis
Macromolecules and polymers, including “molecular engineering”
Self-assembly
Molecular recognition and selective chemical interactions, sensors.
Biology and medicine
Nanosurgery, genetics - including DNA testing - polymerase processes
Applied nanotechnology - an example in solar energy
Our planet - sunlight and water => hydrogen + oxygen => energy!
Ideal! - but to be realistic:
The "Solar Constant” is 1.37 x 10 3 W/m2
So Earth receives 1.2 x 1017 W insolation or
1.56 x 1018 kWh/year in total.
1 kg hydrogen
= 39.4 kWh
So sunlight represents
1016 kg H2
3.9 x
Typical solar cell efficiency is 10% so
midday electric power is 100 W/m2,
Annual energy harvest (N. Europe)
80kWh/m2 , or 2 kg. H2
But 0.13% of earth’s surface covered
with PV panels of 10% efficiency =
present world total energy demand!
It’s a matter of scale and economics!
Present-day PV is too costly - can
nanotechnology provide an alternative?
Power source for solar cells
1.6
nm-1 ]
Solar global normal spectral irradiance, (AM 1.5).
1.4
-2
Solar direct normal spectral irradiance, (AM 1.5).
P=1.367kWm-2 - the
solar constant – solar
radiation power outside
the Earth’s atmosphere
1.0
1.5
(λ
1.2
0.8
Solar diffuse normal spectral irradiance, (AM 1.5).
0.6
0.4
0.2
0.0
500
1000
1500
2000
2500
Wavelength [nm.]
3000
3500
4000
Another bit of history
1839 Edmund Becquerel, a French physicist,
observed the photovoltaic effect in a
photoelectrochemical system. The electrode
was a nanostructure - a thin film of silver
halide on silver!
(E. Becquerel,"Mémoire sur les effets électriques produits
sous l'influence des rayons solaires", C. R. Acad. Sci. Paris,
1839, 9, 561-567)
The effect was first observed in a solid material
(selenium semi-metal) by Willoughby Smith (1873) and
in 1876, William Grylls Adams and Richard Evans Day
discover that selenium produces electricity when
exposed to light, - solid material could change light
into electricity without heat or moving parts.
Semiconductor energy levels and band gap
(band gap in semiconductor = HOMO--LUMO gap in molecular system)
THE PHOTOVOLTAIC CHARACTERISTICS OF SOLAR CELLS
How does the device work?
The p-n junction under illumination (on the right). A photon induced hole-electron pair
is separated by the local field of the junction. Taken from: F. C. TREBLE (Editor); Generating
Electricity from the Sun; Pergamon Press, Inc.;New York; 1991
.
Photoelectrochemical effect at a semiconductor
- redox electrolyte interface.
a) On contact the Fermi level of the n-type
semiconductor equilibrates with that of the
metal and with the redox level of the electrolyte.
After charge (electron) transfer a band bending
is established as in the case of the previous
solid-state junctions, with establishment of the
depletion zone.
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
b) Under light, photoelectrons enter the
conduction band; the band bending is reduced
and a photovoltage is generated between the
semiconductor Fermi level and the redox
potential of the electrolyte - equivalent to the
potential of the metal counter-electrode. Minority
carriers - holes - are then available for an
oxidation reaction with the electrolyte at the
Semiconductor photoanode. A reduction
reaction takes place at the cathodic
counterelectrode.
Semiconductor - electrolyte junctions photoelectrochemistry
The two materials with different conduction mechanisms - may also be a
semiconductor and an electrolyte giving a photoelectrochemical device. The
first observation of a photovoltaic effect (Becquerel, 1839) was in fact a
photoelectrochemical system.
Narrow band-gap semi-conductors
whose photo-response matches the
solar spectrum are in general unstable
in contact with electrolytes. When a
hole reaches the interface the most likely
object of an oxidation reaction is the
material of the semiconductor itself! e.g.
CdS + 2h+ -> Cd2+ + S.
This is an example of photocorrosion.
Wide band-gap semiconductors have
stronger chemical bonding, and are
therefore more stable.
Nature’s solar cell - photosynthesis in
the green leaf
The natural prototype for a
solar energy conversion dye - chlorophyll
There are other systems besides semiconductors
which can absorb visible light and store the
acquired energy. Photosynthesis - based on
chlorophyll - is the primary energy source of the
biosphere.
So organometallic porphyrins are a possibility for
solar energy conversion.
It is also known (Moser, 1887) that dyes can
enhance the photoelectrochemical performance
of semiconductors.
Essentially the same process was developed at
the same time for photography - panchromatic
films and ultimately colour photography.
First report of a dye-sensitised photoeffect on an
illuminated semiconductor - Moser, Vienna, 1887.
Sensitivity to solar spectrum of titanium dioxide
and of electroactive dye N3.
Solar spectrum
Irradiance
Dye N3
IR
Visible
Photon energy
Titanium
oxide
UV
How photosensitisation works
E
eeŠ
eeŠ
cb
(S+/S*)
hν < Eg
hν1
vb
hν2
(S+/S)
X
Charge injection
S*
S+ + e–cb (SC)
Photoelectrochemical processes in a
dye-sensitized solar cell.
Conducting
glass
TiO2
Dye
Cathode
Electrolyte
Injection S*
-0.5
Maximum
Voltage
0
hν
E vs
NHE
0.5
(V )
Red Mediator Ox
Diffusion
1.0
S¡/S+
e
-
e
-
In a molecular system such as the dye, the gap between the highest occupied
molecular orbital and the lowest unoccupied level (HOMO-LUMO gap) is
analogous to the conduction band - valence band gap in a semiconductor.
Light harvesting by nanocrystalline oxide films
What won’t work: sensitisation of bulk wide-gap semiconductors. The
sensitiser molecule must intimately contact the surface if it is to transfer
charge. But a monomolecular dye film will not absorb enough light!
Solution: use films made of a network of undoped (insulating) wide band
gap oxide nanocrystallites. The extended surface area allows sufficient dye
to be adsorbed.
•
For a 10 micron thick oxide film the surface is enlarged over 1000 times
allowing for efficient harvesting of sunlight by the adsorbed monolayer of
sensitizer. The semiconductor is now an engineered nanostructure.
•
The sensitizer molecule can be designed to graft onto the oxide surface
through suitable anchoring groups, e.g. carboxylate, phosphonate or
hydroxamate. The molecule itself becomes a nano-device!
•
Sensitized electron injection from the adsorbed dye renders the oxide
conductive.
•
The circuit is completed through
semiconductor film nano-porosity.
an
electrolyte
penetrating
the
The system is regenerative loss mechanisms are slow & inefficient
s
hν
excitation
+
s
s*
Photoinduced electron transfer mechanism
Photoexcitation:
(S+/ S*)
e–
cb
S/SC + hν
S*/SC
Charge injection:
hν
S*/SC
e– (SC) + S+/SC
Dye regeneration:
e–
e–
e–
(S+/
S)
(D+/ D )
S+/SC + R
S/SC + R+
Charge recombination:
S+/SC + e–
S/SC
Dark current:
Semiconductor
TiO2 - SC
Dye-sensitizer
S
Redox Mediator R+ + e–
- electrolyte R
R
Dynamic Competition
electron
injection
dye
regeneration
electron
interfacial
transport recombination
time [s]
electron transport
loss mechanism:
interfacial
recombination
Competition ⇒
Electron diffusion length
Ln =
Dn ⋅ τ n
τn: electron lifetime
Dn: electron diffusion coefficient
A cross section of the dye sensitized solar cell
e-
Working
electrode
Glass
Fluorine-doped SnO2
TiO2 with monolayer of dye
Redox electrolyte I-/I3Pt Catalyst
Fluorine-doped SnO2
e-
Glass
Counter
electrode
B. O’Regan, M. Grätzel, Nature 1991, 353, 737−740
M. Grätzel, Nature 2001, 414, 338−344.
Some idea of scale!
On the nano scalee
-
3
-
S+/S*
1
Oxidation
Potential
2
e
-
5
+
TiO
2
4
S+/S
Red/Ox
Couple
Standard dye for photoelectrochemical cell development
- the EPFL « N3 » dye dithiocyanato bis(4,4’-dicarboxylic acid-2,2’-bipyridine)
ruthenium(II)
COOH
Note the structure and compare
with chlorophyll - in both cases a
metal in a “cage” of nitrogen atoms,
these being in heterocyclic aromatic
rings.
COON
S
C
N
N
Ru
N
N
C
S
N
COO-
COOH
Absorption spectra of two isomers of the ruthenium dicarboxydipyridyl, dithiocyanate, or N3-type, dye demonstrating sensitivity of
optical properties to molecular structure
0.8
COOH
HOOC
COOH
HOOC
0.6
N
N
N
NCS
N
Ru
SCN
NCS
Ru
HOOC
N
NCS
N
N
N
0.4
HOOC
OOCH
HOOC
0.2
0.0
300
400
500
E
600
[
]
700
800
Stabilisation of the trans- structure,
using a tetradentate polypyridyl ligand
Dye chemistry and colour
COOH
COOH
S
C
HOOC
COOH
HOOC
N
S
N
HOOC
N
N
N
N
N
N
N
N
N
N
N
HOOC
COOH
COOH
HOOC
COOH
N
C
C
N
N
Ru
Ru
Ru
S
N
C
C
S
N
HOOC
COOH
S
Absorption spectra of modified
dye structures
A look at the molecular geometry
COOH anchoring groups
Hydrothermal growth of nanostructured titania
Precursor Preparation
Modify Ti-isopropoxide with acetic acid
Hydrolysis
Rapidly add precursor to water
Peptization
Acidify with HNO
Hydrothermal Growth
Concentrate Colloid
Solvent Exchange,
Ethyl Cellulose Addition
Homogenize Paste
3 , reflux
Autoclave: 12 hours at 230°C
Rotovap: 45°C, 30mbar
Flocculation, centrifuging
3-roll mill, 15 minutes
Screen Print Films
Dry and Anneal Films
Anneal: 450°C, 20 minutes
Titanium dioxide (anatase) nanocrystals
40
30
20
100 nm
Preferred (101) orientation
of surface planes visible
10
0
0
10
20
30
40
Diameter of particles, nm
50
Dye chemisorption on titanium dioxide
7.2 Å
4.8 Å
Thickness of one dye molecular monolayer = 1 nm!
A further functionalisation modified hydrophobicity
O
O
O
O
O
O
O
N
O
N
N
C
S
Ru
N
N
K51
O
N
OH
N
N
Z907
C
S
N
C
S
Ru
N
N
N
O
C
S
OH
NaO
O
NaO
O
Chemical structure of the hydrophobicity-modifying sensitising dyes,
K51 (tetraethylene oxide side chains) and Z907Na (nonyl side chains);
this inhibits loss of dye by hydrolysis of the attaching bonds.
Design features of the EPFL dye Z-907
1. Chromophore provided by polypyridyl complex of ruthenium the prototype dye of this series is the trisbipyridyl compound.
2. Energetics - HOMO-LUMO gap and hence spectral response modified by substitution of thiocyanide groups.
3. Chemisorption to titanium dioxide surface through carboxylate
groups.
4. Layer self-organisation and hydrophobic surface characteristics
determined by hydrocarbon « tail » with suppression of redox
capture of injected electrons by the electrolyte.
5. Particularly suitable with ionic liquid and gel electrolytes.
A brief mention of the electrolytes
1-Ethyl-3-methylimidazolium tetracyanoborate (EMIB(CN)4)
is a new ionic liquid of a low viscosity (19.8 cP at 20 ｰC)
and high chemical and thermal stability.
Objective: wider thermal tolerance, above 200ºC
Characteristics of a dye-sensitised cell under light
Efficiency 11.04%
100% AM1.5
•
Current Density[mA/cm
2
]
-15
Efficiency 11.18%
65% AM1.5
•
-10
-5
Efficiency 10.87%
9.5% AM1.5
•
0
0.0
0.2
0.4
Potential [V]
0.6
0.8
Adding beauty to function
80%
60
IPCE [%]
WMC 273
40
N
N
Zn
N
N
20
CO2 H
HO2 C
0
400
500
600
Wavelength [nm]
700
800m
Sensitisation with a zinc phthalocyanine.
Colour variation in a
series-connected DSC module
Courtesy Dr. Winfried Hoffman, CEO, RWE, SCHOTT Solar GmbH
Product concept - flexible cells
(Konarka Inc., USA)
Product concept - flexible cells
(Hitachi-Maxell, Japan)
Outdoor installation CSIRO, Newcastle NSW, Australia
(Dyesol Ltd., Australia)
Another possibility - Electrochromic display
coloration
e-
cb edge link
electrochromophore
SnO2
potential
Ntera Ltd.,Ireland
potential
How it works
SnO2
discoloration
TiO2
CTO
viologen modified
TiO2 film
phenothiazine modified
SnO2:Sb film
link
electrochromophore
e-
cb edge
TiO2
Adsorber
Molecular
Monolayer
SnO2
HO
+
N
O P
O
2 nm
Seriseal
Electrode
20 nm
+
N
ÉCOLE POLYTECHNIQUE
FÉDÉRALE DE LAUSANNE
Use of the AM 0 solar spectrum by a silicon solar cell
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.