A cartoon model of a single‐threshold solar cell. Low energy photons (red) cannot excite electrons to the threshold required to create electrical energy. Only yellow and blue are utilised here. Public Dissemination Report A‐023 Upconversion of the solar spectrum for improved PV energy conversion This project was funded by ARENA, to investigate how the process of photochemical upconversion could be applied to solar cells, in order to boost their light harvesting efficiency. The project was largely based at The University of Sydney, with contributions from Helmholtz Zentrum Berlin, Imperial College, London, The University of Adelaide, and UNSW. The project succeeded in demonstrating upconversion applied to amorphous silicon solar cells, and other types such as organic solar cells and dye‐sensitized solar cells. The project also explored ways to make upconversion more efficient: using plasmons and polymers. Single threshold solar cells The solar cells that one sees going about everyday activities are made of silicon, and have a single energy threshold. Such cells do not absorb particles of light (photons) with an energy below about 1.1 electron‐ volts (eV, the energy it takes to move an electron across a voltage of 1.1 volts). The entire visible spectrum spans 1.6‐3.0 eV, and so silicon solar cells can make use of these photons, and part of the infrared spectrum. However, whether they absorb a red (2 eV) or blue (3 eV) photon, they will only use 1.1 eV of the energy, their energy threshold, or band‐gap. These limitations, missing the photons below the band‐gap, and wasting the portion of energy above the band‐gap, limit the energy conversion efficiency of a single‐ threshold silicon solar cell to about 30%. Australian researchers have pushed silicon to exceed 25% efficiency, which approaches the fundamental limitation – the silicon ceiling. To push through the silicon ceiling, one requires better usage of the solar spectrum. Why is efficiency important? The cost of solar energy is the ultimate driver for its adoption by the wider community. For a roof‐ mounted system with a limited area, more energy can be extracted if the solar modules themselves are more efficient. Since much of the cost of an installation is fixed, whether the modules are efficient or not, increased efficiency is the easiest way to drive down the cost per kilowatt of an installation.
11
Embed
Public Dissemination Report - Australian Renewable Energy ... · EQE, transm, rubrene em ... It was designated as a “hot article” by the publisher, the Royal Society of Chemistry,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A cartoon model of a single‐threshold solar cell. Low energy photons (red)
cannot excite electrons to the threshold required to create electrical
energy. Only yellow and blue are utilised here.
Public Dissemination Report
A‐023 Upconversion of the solar spectrum for improved PV energy conversion
This project was funded by ARENA, to investigate how the process of photochemical upconversion
could be applied to solar cells, in order to boost their light harvesting efficiency. The project was
largely based at The University of Sydney, with contributions from Helmholtz Zentrum Berlin,
Imperial College, London, The University of Adelaide, and UNSW.
The project succeeded in demonstrating upconversion applied to amorphous silicon solar cells, and
other types such as organic solar cells and dye‐sensitized solar cells. The project also explored ways
to make upconversion more efficient: using plasmons and polymers.
Single threshold solar cells
The solar cells that one sees going
about everyday activities are made
of silicon, and have a single energy
threshold. Such cells do not absorb
particles of light (photons) with an
energy below about 1.1 electron‐
volts (eV, the energy it takes to
move an electron across a voltage
of 1.1 volts). The entire visible
spectrum spans 1.6‐3.0 eV, and so
silicon solar cells can make use of
these photons, and part of the
infrared spectrum. However,
whether they absorb a red (2 eV) or blue (3 eV) photon, they will only use 1.1 eV of the energy, their
energy threshold, or band‐gap. These limitations, missing the photons below the band‐gap, and
wasting the portion of energy above the band‐gap, limit the energy conversion efficiency of a single‐
threshold silicon solar cell to about 30%.
Australian researchers have pushed silicon to exceed 25% efficiency, which approaches the
fundamental limitation – the silicon ceiling. To push through the silicon ceiling, one requires better
usage of the solar spectrum.
Why is efficiency important?
The cost of solar energy is the ultimate driver for its adoption by the wider community. For a roof‐
mounted system with a limited area, more energy can be extracted if the solar modules themselves
are more efficient. Since much of the cost of an installation is fixed, whether the modules are
efficient or not, increased efficiency is the easiest way to drive down the cost per kilowatt of an
installation.
In upconversion, two low energy
photons work together to excite
electrons by converting into higher
energy photons.
(a) The energy flow in photochemical upconversion. (b) Typical sensitizer molecules (left) and emitter molecules
(right) used in this work.
ISC ISCS1
S0
S1
S0
T1 T1
S1
S0
S1
T1 T1
sensitizer emitter emitter sensitizer
hν2
hν1hν1
TET TET
TTA
N
N
N
N
NN
N N
N
NN
NPd
R1
R , R = HPQ Pd: rubrene1
R = NH2
24
PQ PdNA:4 2R = NO ,1 2
R2
(a) (b)
How can we make better use of the solar spectrum?
Multiple threshold cells, which have several bandgaps to make good use of the energy of the solar
spectrum, are only found in applications where the cost of the cells is not the primary consideration,
such as spacecraft. They are incredibly expensive. However, all cells of the single‐threshold type
naturally transmit photons of lower energy than their bandgap. If this light can be converted into a
usable energy, then the efficiency of the solar cell can be increased. The process whereby two or
more photons of low energy are “glued together”, in effect, to generate photons of higher energy, is
called upconversion.
What is upconversion?
Upconversion requires absorption of low energy photons by
some material, with the later interaction of two or more
quanta of the stored energy to liberate higher energy
photons. While this has been known to occur in inorganic
materials, such as rare earth phosphors, for some time, the
efficiency is low.
In photochemical upconversion, two types of organic
molecule are used. One type, the sensitizer, absorbs the low
energy photons and stores the energy as a metastable state.
The other type, the emitter, harvests energy from the
sensitizers. Two emitters in metastable states interect to
generate a highly emissive electronic state. The emitters
subsequently emit energy derived from two absorbed low
energy photons. The emitted energy is above the band‐gap of
the solar cell to which it is coupled. Thus, unused photons are
conjoined to generate usable photons. This boosts the
efficiency of the solar cell.
Project Aims
Before the commencement of this project in 2011, there were no reports of photochemical
upconversion being applied to solar cells. The goals of this project were several‐fold: to apply
photochemical upconversion to a solar cell for the first time; improve the efficiency of the
upconversion process by increasing the absorption bandwidth of the sensitizers; explore polymers as
emitting materials; and apply plasmonic nanoantennas to improve light harvesting.
External quantum efficiencies of the a‐Si:H solar cells,
compared to absorption of sensitizer (), and emission of
emitter. Cells have a diminished response, and enhanced
transmission in the red part of the spectrum (>650nm).
(a)
(b)1.0
0.8
0.6
0.4
0.2
0.0
EQ
E, transm
, rubrene em
3.0 2.5 2.0 1.8
700600500400wavelength (nm)
4x10-16
3
2
1
0
σ (c
m2 )
1.0
0.8
0.6
0.4
0.2
0.0
EQ
E, transm
, rubrene em
3.0 2.5 2.0 1.8photon energy (eV)
700600500400
4x10-16
3
2
1
0
σ (c
m2 )
The efficiency increases observed for the two sensitizers
employed.
2.0
1.5
1.0
0.5
0.0rel.
effic
ienc
y in
crea
se (%
)
750700650600550
2.2 2.0 1.9 1.8 1.7photon energy (eV)
1.5
1.0
0.5
0.0
rel.
effic
ienc
y in
crea
se (%
)
750700650600550wavelength (nm)
2.2 2.0 1.9 1.8 1.7
(a)
(b)
Application of photochemical upconversion to a solar cell for the first time
Hydrogenated amorphous silicon solar cells
were manufactured for the project by
Helmholtz Zentrum Berlin, under the direction
of Prof. Dr. Klaus Lips and Dr. Bernd
Stannowski. As shown, they have diminished
response in the red region of the spectrum,
which is complemented by the absorption cross
section of the sensitizers chosen. A broadband
sensitizer (b) was developed by Dr Tony Khoury
especially for this project.
A liquid upconvertor was optically contacted to
the rear of the solar cell using immersion oil.
The oil minimized losses of light from scattering
off interfaces.
External quantum efficiency (EQE) curve
measurements were made while biasing the
upconvertor with a known amount of light. This
was required since the upconvertor needs a
certain amount of light to work at its full
potential. Indeed,
The measurement was repeated with the
system slightly misaligned. The ratio of the two
EQE curves demonstrates the extra EQE made
available to the solar cell due to upconversion.
The EQE was increased by about 1‐2% at the
peak absorption of the sensitizer. This
increased EQE can be expressed as the
expected (short circuit) current increase under
standard illumination. This Figure‐of‐Merit was
calculated to be 2.8×10‐5 mA.cm‐2 and 1.3×10‐4
mA.cm‐2 for the narrow and broadband
sensitizer, respectively.
These figures are some 1000 times lower than
what is device‐relevant. Nevertheless,
compared to state‐of‐the‐art measurements
using inorganic phosphors, photochemical
upconversion was shown to be 200 times more
efficient.
n (a-Si, 20 nm)i (a-Si:H, 50 nm)p (a-SiC, 10 nm)
immersion oil
ZnO:Al (300 nm)
TCO (SnO2:F, 800 nm)
glass (3.2 mm)
quartz cuvette (1.25 mm)
quartz cuvette (1.25 mm)
UC solution (1 cm)
SC
UCsilver coated glass spheres (100μm)
Position 1
Position 2
This work was published in the high‐profile journal Energy and Environmental Science.
Improving the light‐harvesting of amorphous silicon solar cells with photochemical upconversion, Y. Y.
Cheng, B. Fuckel, R. W. MacQueen, T. Khoury, R. G. C. R. Clady, T. F. Schulze, N. J. Ekins‐Daukes, M. J.
Crossley, B. Stannowski, K. Lips and T. W. Schmidt, Energy & Environmental Science, 5, 6953‐6959,
(2012)
It was designated as a “hot article” by the publisher, the Royal Society of Chemistry, and has been
cited over 80 times already.
Increasing device fidelity using a back‐reflector
Photochemical upconversion is necessarily an
isotropic process. That is to say, the upconverted
light is emitted in all directions. As such, half of
the upconverted light is emitted away from the
solar cell. To remedy this, we created a scattering
back‐reflector using 100m silver‐coated beads.
The EQE curves were measured as before, but at
two positions on the upconvertor cuvette: with
and without silver‐coated beads.
The result was a doubling of the observed Figure‐
of‐Merit (FoM). At the same time, adjustments
were made to the upconvertor solution, which
also increased the FoM.
With these improvements, the FoM was lifted to
3.4×10‐4 mA.cm‐2.
This work was published in The Australian Journal
of Chemistry, in a special issue commemorating
the Conference of the Physical Chemistry Division
of the RACI in late 2011, where Schmidt presented
a Keynote address.
Photochemical Upconversion Enhanced Solar
Cells: Effect of a Back Reflector, T. F. Schulze,
Y. Y. Cheng, B. Fuckel, R. W. MacQueen, A.
Danos, N. J. L. K. Davis, M. J. Y. Tayebjee, T.
Khoury, R. G. C. R. Clady, N. J. Ekins‐Daukes,
M. J. Crossley, B. Stannowski, K. Lips and T. W.
Schmidt, Australian Journal of Chemistry, 65,
480‐485, (2012)
The improved FoM shows that enhancements can be made to the device by engineering both at the
molecular level, and at the level of optical device architecture.
Embossed back‐reflectors improved
upconversion by more than a factor of 2.
10-3 10-2 10-1 1000.0
0.5
1.0
1.5
2.0
2.5
Experiment:embossedflat
Ray tracing:embossedflat
(a)
rel.intensity
ofUCsig
nalinsolarcellEQE
thickness of upconverting layer (mm)
100 µm
20 µm
20 µm
40 µm
50 µm
Improved EQEs of organic (top two) and a‐Si:H
solar cells.
Micro‐Engineered Back‐Reflector
Since upconversion is a 2‐photon process, it is more efficient if
light is concentrated by a lens or mirror. Taking the back‐
reflector concept a step further, we created a back‐reflector
with embossed spherical indentations, designed to
concentrate reflected light.
Measurements were undertaken as a function of the distance
of the backreflector to the front of the upconvertor. The peak
was observed at a thickness of about 10m, where it was
found that the embossed back‐reflector improved
upconversion by more than a factor of 2 as compared to
without a back‐reflector.
This work was published in the Journal of Photonics for Energy:
Micro‐optical design of photochemical upconverters for thin‐
film solar cells, T. F. Schulze, Y.Y. Cheng, T. Khoury, M. J.
Crossley, B. Stannowski, K. Lips and T. W. Schmidt, Journal
of Photonics for Energy, 3, 034598 (2013)
Improved Device matching and application to organic
photovoltaics
While it is, in principle, possible to manipulate the
chemical structure of sensitizer molecules to alter its
spectral properties, in practice this is a laborious and
lengthy task. In order to demonstrate an improved FoM
by matching a solar cell to the upconvertor, bespoke a‐
Si:H cells were manufactured by HZB. Organic solar cells
were obtained from Karlsruhe Institute of Technology.
An improved FoM was obtained for the a‐Si:H cells, of
7.6×10‐4 mA.cm‐2. EQEs of organic cells were improved
by 12% peak.
This work was published in Journal of Physical Chemistry C:
Efficiency Enhancement of Organic and Thin‐Film Silicon Solar Cells with Photochemical Upconversion,
T. F. Schulze, J. Czolk, Y. Y. Cheng, B. Fückel, R. W. MacQueen, T. Khoury, M. J. Crossley, B. Stannowski,
K. Lips, U. Lemmer, A. Colsmann, T. W. Schmidt, The Journal of Physical Chemistry C, 116, 22794–
22801 (2012)
Gold nanoparticles embedded within a dielectric
layer such as silica can be decorated with
sensitizer molecules, and transfer this energy to
an emitter material.
hν1
hν1
hν2
ΤΕΤΤΕΤ
ΤΤΑ
Plasmonic Antennas for Upconversion
In order to improve the light‐harvesting characteristics of sensitizer molecules, we proposed to
utilize plasmonic nanoparticles as antennas. The cooperative oscillations of electrons in small metal
particles has a characteristic frequency which depends on the size of the nanoparticles.
The most famous demonstration is the scattering
of green light by plasmonic gold nanoparticles in
the Lycurgus Cup, a 4th century Roman artwork.
The cup appears red if illuminated from the
inside, due to resonant scattering of green. In
ambient light, it scatters the green back to the
eye of the observer. The resonant frequency of
the antennas can be tuned to resonate with the
absorption frequency of sensitizer molecules.
In collaboration with Professor Stefan Maier and Dr Ned Ekins‐Daukes, of Imperial College, London,
we attempted to apply plasmonic light harvesting to upconversion. In our concept, sensitizer
molecules in the vicinity of a bespoke, 2d plasmonic substrate will draw energy from the
nanostructure.
Roland Piper, from Imperial College, London, a PhD student under the supervision of Dr Ned Ekins‐
Daukes, visited in March and April 2012. He brought with him the plasmonic substrates produced by
electron beam lithography. We were unsuccessful in determining any discernable effect of the
plasmonic substrate. We reason that this is because the effect of the plasmonic device is limited to
the ~200nm vicinity, but at the concentrations of
sensitizer material employed, our characteristic
absorption length is more like 100m. As such, we
require a thousandfold increase in absorption
coefficient. This is to be achieved with 3d
nanostructured absorbers currently under investigation.
Furthermore, we will incorporate plasmonic particles in
the centre of the nanosphere supports to effect the
plasmonic enhancement desired.
The concepts under investigation were reported by us
in a book chapter:
CHAPTER 15, Triplet–triplet Annihilation Up‐conversion, Timothy W. Schmidt and Murad J. Y.
Tayebjee, Advanced Concepts in Photovoltaics, 2014, 489‐505, DOI:10.1039/9781849739955‐00489
Poly pendant perylene was synthesized. It was
found to upconvert in solution, but not in the
solid.
Polymer Emitters for Upconversion
The rate of triplet‐triplet annihilation and thus the efficiency is limited by the diffusive motion of
emitter molecules bearing triplet states. It was our concept that the electronic state, rather than the
molecule diffuse. This could be achieved using polymer emitters, where the electronic excitation
jumps along and between chains.
We synthesized a poly‐pendant perylene polymer, but
it was found to upconvert only as a solution, and not as
a solid, presumably because low energy sites have the
effect of trapping the triplet excited states.
While this was unsuccessful, we determined that a
solid phase material could be achieved by another
route which is currently under investigation.
Liquid upconvertors can be rendered solid by the
application of gelators. This work will be published
shortly.
Insights from computer simulations
High performance computers can be used to simulate the movements of molecules in solution, such
as upconverter emitters, in great detail. Led by Dr David Huang and Simon Blacket at the University
of Adelaide, these kinds of simulations have helped support and direct efforts to optimise
upconverter efficiency, and expanded understanding of the upconversion process.
When pairs of emitter molecules interact, their relative positions and orientations in space
determine whether the interaction will be successful and result in upconversion. Prior to these
simulations it was widely assumed that emitter molecules come together in random orientations,
and that as a result, successful interactions were rare.
Instead, the shapes of the emitter molecules strongly influence the molecule’s interactions, and
from simulations the preferred interaction geometry can be identified. Knowing these preferred
geometries has led to a better understanding of why some emitters are intrinsically more efficient
upconverters than others: the shape of the emitters can cause them to naturally come together in
geometries which promote successful interactions. This work also gives hints to why upconverters
with several different emitters working together can perform better than those with just one.
People
The preferred interaction geometry of rubrene molecules from top‐down (left) and side‐on (right)