Characterising Dye-Sensitized Solar Cells

Laura L. Tobina,, Thomas O’Reillyb,, Dominic Zerullab, John T. Sheridana*

a
SFI-Strategic Research Cluster in Solar Energy Conversion,
UCD Communications and Optoelectronic Research Centre,
School of Electrical, Electronic and Mechanical Engineering,
College of Engineering, Mathematics and Physical Sciences,
University College Dublin,
National University of Ireland,
Belfield, Dublin 4, Ireland.;
b
School of Physics,
College of Engineering, Mathematics and Physical Sciences,
University College Dublin,
National University of Ireland,
Belfield, Dublin 4, Ireland.

ABSTRACT

With growing energy and environmental concerns due to fossil fuel depletion and global warming there is an increasing
attention being attracted by alternative and/or renewable sources of power such as biomass, hydropower, geothermal,
wind and solar energy. In today’s society there is a vast and in many cases not fully appreciated dependence on
electrical power for everyday life and therefore devices such as PV cells are of enormous importance. The more widely
used and commercially available silicon (semiconductor) based cells currently have the greatest efficiencies, however the
manufacturing of these cells is complex and costly due to the cost and difficulty of producing and processing pure
silicon. One new direction being explored is the development of dye-sensitised solar cells (DSSC). The SFI Strategic
Research Centre for Solar Energy Conversion is a new research cluster based in Ireland, formed with the express
intention of bringing together industry and academia to produce renewable energy solutions. Our specific area of
research is in biomimetic dye sensitised solar cells and their electrical properties. We are currently working to develop
test equipment, and optoelectronic models describing the performance and behaviors of dye-sensitised solar cells
(Grätzel Cells). In this paper we describe some of the background to our work and also some of our initial experimental
results. Based on these results we intend to characterise the opto-electrical properties and bulk characteristics of simple
dye-sensitised solar cells and then to proceed to test new cell compositions.

Keywords: Solar, dye-sensitised, biomimetic, Grätzel Cells

1. INTRODUCTION

Conventional silicon and more exotic quantum based solar cells continue to drive the solar energy production and
dominate the commercial market. However over the last decade momentum has grown in support of organic solar cell
technology. Silicon solar cells flourished over the years due to the knowledge and material obtained from the
semiconductor industry because of the technical expertise and financial investment in fabrication plants in these areas,
conventional PV cells have acquired respectful efficiencies1.

* [Link]@[Link]; phone: +353 1 716 1927

1
The disadvantage of silicon solar cells is that the associated manufacturing process is complex and costly2. While it
would seem that there is a plentiful supply of the required raw material, since after oxygen, silicon is the second most
abundant element and comprises of 25.7% of the Earth’s crust. The disadvantage of working with silicon is that it is not
found free in nature but is found as oxides (e.g. sand, quartz, flint) and as silicates (e.g. granite, asbestos, clay). Energy
intensive processing and clean room fabrication condition are essential when working with this material3.

One economically viable and flexible alternative to the p – n junction photovoltaic (PV) devices is to look at thin film
organic-material devices, specifically dye-sensitised solar cells DSSC. In silicon systems the semiconductor takes on the
dual role of charge carrier transport and light absorption, whereas in DSSC these two operations are separated4.

An attractive feature of these dye-sensitised solar cells is that the concept derives from the area of biomimetics, that is
that DSSC devices are made so as to imitate or mimic nature in some way. In this case there are analogies to the process
of photosynthesis. In both cases incoming light is absorbed by an organic dye and electrons are produced, resulting in
the production of positive and negative charge carriers.

A schematic of a DSSC can be seen in Figure 1 where a monolayer of an organic dye which is attached to a mesoscopic
film of a wide bandgap oxide semiconductor. The dye absorbs incident light and produces electrons, which are injected
into the conduction band of the semiconductor oxide. The electrons are transported across the nanoparticle layer to the
current collector (anode). From here the electron passes through the external circuit and subsequently to the counter
electrode (cathode). The sensitiser is regenerated by the organic hole conductor and transports the positive charges to the
counter electrode where recombination occurs. Under solar exposure there is no net chemical change; light is converted
to electrical power.

Light

Glass and transparent conducting

e- e-
e- coating

Dye

TiO 2
Load

I3- Electrolyte
3I-

Catalyst
-
e
e- e- Glass and transparent conducting
coating

Figure 1. Schematic of the energy flow in a dye sensitised solar cell.

2
Both natural photosynthesis and DSSCs use organic dyes such as anthcyanins. In DSSCs a mesoscopic film of titanium
dioxide replaces nicotinamide adenine dinucleotide phosphate (NADP+), and carbon dioxide replaces the electron
acceptor. Iodide and triode (I-, I 3 -) replaces water and oxygen as the electron donor and oxidation product5.

Operational devices can be made in non-clean room conditions and can be made reasonably robust to environmental
conditions unlike silicon solar cells.

2. HISTORICAL OVERVIEW

Edmund Becquerel is credited with being the first person to have reported the photoelectric effect in 1839. He noted that
a photocurrent was produced when silver coated platinum electrodes were immersed in electrolyte6,7.

The next significant development came from Willoughby Smith’s experiments in 1873 on electrical currents passing
though a bar of crystalline selenium, which found that its resistance was reduced when the bar of crystalline selenium
was exposed to light rather than been kept in the dark. Following on from this in 1897 William Adams and Richard Day
reported that a current was produced when light fell upon selenium, which had two heated platinum contacts and no
external power supply8.

In 1894 Charles Fritts constructed the first large area solar cell by using plates made from two different metals and
compressing molten selenium between them. Fritts was able to adhere a thin semi-transparent layer of gold leaf onto a
thin film of selenium to produce the first thin film photovoltaic. The results of this experiment paved the way for the
study of solar cells in the 19th century9, 10.

The photoelectric effect was also detected in copper-cuprous oxide thin film structures, in lead sulphide and thallium
sulphide. In these cells there was a layer of semitransparent metal deposited onto the semiconductor, which provided the
asymmetric electronic junction11. It was Goldman and Brodsky in 1914 that associated the photoelectric effect to the
existence of a barrier to current flow at one of the interfaces of the semiconductor metal, i.e., rectifying action.
Throughout the 1930’s Neville Mott and Walter Schottky predominantly pioneered the development of a theory of metal
semiconductor barrier layers.

The development of silicon electronics followed the innovation of a technique to manufacture p – n junctions in silicon
in the 1950’s. The p – n junction structure architecture produced superior rectifying action and photovoltaic
performance1.

Prior to 1953 the most efficient photovoltaic devices were selenium photocells with a maximum efficiency of 0.8 %.
This changed dramatically in 1954 when D. M. Chapin, C. S. Fuller and G. L. Pearson combined their research to
produce the first silicon solar cell with an efficiency of 6%. Other materials such as gallium arsenide, indium phosphide
and cadmium telluride were studied for use in p – n junction photovoltaic devices, but silicon has remained the principal
photovoltaic material. Silicon has benefitted form the advances of silicon technology from the microelectronics
industry12, 13.

In the 1970’s alternative energy sources research was spurred on by the energy crisis in the oil dependent western world.
There was a growth in funding for research and development of photovoltaics and a variety of methods for reducing
manufacturing and material costs while simultaneously improving device efficiency. Photochemical junctions were
explored with a view to lower costs, and alternative materials included amorphous silicon, polycrystalline silicon and
organic conductors. To improve efficiencies, tandem and other multiple band gap structures were developed1.

3
With the discovery of the hole in the Ozone layer in 1985 there was a renewed wave of pressing concern for cleaner
energy alternatives. In 1991 Brian O’Regan and Michael Grätzel reported a new concept of a solar cell that was based
on biomimetics, specifically mimicking the photosynthesis process in plants4. This type of solar cell is known as a dye
sensitised solar cell. To date their efficiencies are not a current threat to silicon solar cells, however it is hoped that one
day these dye-sensitised solar cells will improve significantly to be considered for commercial use.

Photovoltaic production has expanded at a rate of 15 – 25% per annum since the latter half of the 1990’s and this has
allowed a reduction in manufacturing costs1. Photovoltaics are finally starting to become competitive energy suppliers
as conventional electricity supplies have become more expensive. One example of growth is in commercial solar panels
for residential houses.

3. METHODOLOGY

We have implemented rudimentary dye-sensitised solar cells and performed measurements to characterise the current-
voltage (I-V) values obtained. The dye-sensitised solar cells consist of two glass slides (4 cm  2 cm), which have a
layer of transparent conducting coating on one face of each slide as shown in Figure 1. Fluorine doped SnO 2 was our
material of choice for the transparent conducting coating. Other oxides such as ZnO and SnO 2 can and have been
previously investigated14.

By measuring the electrical conductivity to find the resistance of the glass slides, the transparent conducting coating was
immediately distinguishable from the non-conducting face.

One basic method that was used for applying a monolayer of TiO 2 was masking three sides of the conductive face of the
glass slide with Scotch tape. This formed a mould in which the TiO 2 solution could flow into. Another reason for
masking the conductive glass was to allow for electrical contacts. Using a pipette, several drops of the commercial
colloidal TiO 2 was transferred onto the slide and spread over the unmasked area evenly.

Once the TiO 2 dried the mask was removed with care. Sintering is required to adhere the TiO 2 layer onto the glass slide.
Using a furnace at approximately 450°C the sintering time is roughly 5 – 10 minutes. The TiO 2 layer turns into a
brownish colour, which is due to the organic matter in the TiO 2 layer reverting back to its former white colour once it
has been successfully sintered on. The glass slide has to be cooled slowly after the sintering process to avoid thermal
stress. This is the negative electrode.

One option is to dye the negative electrode. In this experiment dried Hibiscus flowers were soaked in a petri dish filled
with boiling water. The slides were immersed face up in the solution for approximately 10 minutes to ensure that the dye
had completely penetrated the TiO2 layer. This was obvious by the purple-red staining of the TiO 2 layer. Tweezers
were used to remove the slides from the petri dish and cold water was used to gently rinse off any excess dye. The slides
were left face up to dry.

The counter electrode was formed by coating a light carbon layer over the transparent conducting coating of the second
glass slide. Using a pencil, a graphite coating was applied to the surface of the transparent conducting coating and loose
graphite particles were brushed away. This layer acts as the catalyst for the triiodide – iodide (I 3 -, I-) regeneration
reaction.

The positive and negative electrodes are placed together so that the catalyst-coating electrode is placed on top of the
TiO 2 layer. The two glass slides are offset as to allow the entire TiO 2 layer to be covered by the counter electrode and
the bare side of the transparent conducting coating is exposed. Binder clips were used to hold the glass slides together.
To activate the cell the electrolyte solution (triiodide – iodide) was placed at the edge of the glass slides and capillary
action drew the solution down to stain the TiO 2 layer. The exposed sides of the cell were used as the contact points5.

4
 4. RESULTS

For this solar cell dried Hibiscus flowers were used for the dye and a carbon coating was used for the counter electrode.
Other cyanin dyes were explored such as raspberry and blackberry juice, however the observed current values were far
lower than that of the Hibiscus flowers. Using a 20 W 12 V halogen lamp as a light source, set 20 cm from the solar cell
and using the circuit described in Figure 2, the current and voltage can be measured.

I
- 10k Potentiometer
Solar Cell
V
+
Figure 2. Circuit schematic for measuring the current-voltage characteristics of the solar cell.

This circuit comprises a 10 K potentiometer, a voltmeter, an ammeter and a solar cell under test. By varying the
potentiometer and exploiting the relationship between current and voltage from Ohm’s Law a current-voltage curve is
obtained as illustrated in Figure 3.

V  IR .
(4.1)

Figure 3. Typical current voltage (I-V) curve for a solar cell stained with dried Hibiscus flowers.

5
From the knee of the curve, the maximum product of current and voltage was obtained. This value can be used to
calculate the efficiency of the solar cell. The power density of the cell is given by the product of the photocurrent
density by the voltage,
P  JV .
(4.2)

The solar cell will reach a maximum power point, P max , which will occur at a maximum voltage, V m , with a
corresponding current density, J m , which can be seen graphically in Figure 3. The fill factor is the ratio of the maximum
obtainable power to the theoretical power.
J mVm
FF  .
J scVoc
(4.3)

The conversion efficiency, η, of the solar cell can be calculated by measuring fill factor, FF, the power density, P, and
the photocurrent density, J, at short circuit, J sc , and the open circuit photovoltage, V oc .

J scVoc FF
 .
Ps
(4.4)

The key quantities J sc , V oc , FF and η of a solar cell are the crucial performance characteristics. When testing a solar cell
all four quantities must be defined for a particular illumination. The Standard Test Conditions (STC) for solar cells are
at a temperature of 25°C, at a power density 1000 W m-2 in the Air Mass 1.5 spectrum1.

The fill factor was deternine to be 0.542 where J m = 0.26 mA, V m = 266 mV, Jsc – 0.29 mA and Voc = 389 mV.

5. DISCUSSION AND CONCLUSION

The results for the presented DSSCs were not measured under ideal testing conditions. A basic and inexpensive DSSC is
capable of producing a small photocurrent. However the efficiencies of these DSSCs are far below 1%. At the time of
testing the solar cells, a solar simulator was unavailable and an uncalibrated light source was used. Therefore the power
conversion efficiency could not be accurately calculated and hence it was omitted.

Altering the materials in the DSSC can have a significant and positive effect, although the downside is that there is a rise
in production cost. An example of upgrading the DSSC would be to substitute the graphite catalyst coating for a
platinum coating. Another alternative for a considerate increase in efficiency would be to use a ruthenium dye.
Ruthenium dyes are currently the best dyes for DSSCs14. Another feature is the sealing of the DSSCs. The DSSCs used
were not sealed. By implementing these alternatives the efficiency of a DSSC can dramatically improve to
approximately 10 – 11%15, 16.

Hibiscus can be used as a suitable and inexpensive dye for dye sensitized solar cells5. One of the most notably dyes used
in dye sensitized solar cells with high power conversion efficiencies is ruthenium15,16. The disadvantage of ruthenium is
that it is costly and for our measurements hibiscus was a more suitable dye as it was easily sourced and inexpensive.
Another organic dye which will be examined is porphyrin.

In an effort to characterise the DSSC an automated testing rig will be designed in which multiple solar cells will be
tested under the same light source in the same time frame. Various dyes, different cell architectures and cell longevity
will all be examined. This will allow for a quick comparison of different solar cells.

6
 ACKNOWLEDGEMENTS

We acknowledge and thank Science Foundation Ireland for their funding of this Strategic Research Cluster in Advanced
Biomimetic Materials for Solar Energy Conversion and our Industry Partners: Airtricity, Celtic Catalysts Ltd and
SolarPrint Ltd for their support of this Programme.

REFERENCES
[1]
Jenny Nelson, [The Physics of Solar Cells], Imperial College Press, 2003.
[2]
Mauk, M.G., “Silicon solar cells: physical metallurgy principles”, JOM 55 (5), 38 – 42 (2003).
[3]
Los Alamos National Laboratory's Chemistry Division, [Link]
[4]
O’Regan, B., and Grätzel, M., “A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO 2 films”,
Nature 353, 737 – 739 (1991).
[5]
Smestad, G.P., and Grätzel, M., “Demonstrating electron transfer and nanotechnology: A natural dye-sensitized
nanocrystalline energy converter”, J. Chem. Edu. 75 (6), 752 – 756 (1998).
[6]
Becquerel, A.E., “Recherches sur les effets de la Radiation chimique de la lumiere solaire au moyen des courants
electriques”, Comptes Rendus de L’Academie des Sciences 9, 145 – 148 (1939).
[7]
Becquerel, A.E., “Mémoire sur les effects électriques produits sous l’influence des rayons solaires”, Comptes
Rendus de L’Academie des Sciences, 9 561 – 567 (1939).
[8]
Adams, G.E., and Day, R.E., “The Action of Light on Selenium”, Proc. Royal. Soc. (London) 167, 313 – 349
(1847).
[9]
Fritts, C.E., “On a new form of selenium photocell”, Proc. American Assoc. for the Advancement of Science 33, 97
(1883).
[10]
Perlin, John, [From space to Earth], Earthscan, 15 – 24 (1999).
[11]
Grondahl, L.O., “The Copper-Cuprous-Oxide Rectifier and Photoelectric Cell”, Rev. Modern. Phys. 3, 141 – 168
(1933).
[12]
Green, M.A., “Photovoltaics: Coming of age”, Conf. Record 21st IEEE Photovoltaic Specialist Conf. 1, 1- 8 (1990).
[13]
Smits, F.M., “History of silicon solar cells”, IEEE Trans. Electron Devices 23 (7), 640 – 643 (1976).
[14]
Grätzel, M., “Dye-sensitized solid-state hetrojunction solar cells”, MRS Bulletin 30, 23 – 27 (2005).
[15]
Grätzel, M., “Conversion of sunlight to electric power by Nanocrystalline dye-sensitized solar cells”, J. Photochem.
Photobiol A: Chem.,164, 3 – 14 (2004).
[16]
Wang, Q., Ito, S., Gratzel, M., Fabregat-Santiago, F., Mora-Seró, I., Bisquert, J., Bessho, T. and Imai,
H.,“Characteristics of high efficiency dye-sensitized solar cells”, J. Phys. Chem. B 110, 252110 – 25221 (2006).