Dissert
Dissert
Dissert
By
Copyright 2012
by
Nurul Atiqah Binti Mohd Sargawi, 2012
CERTIFICATION OF APPROVAL
by
Approved:
__________________________
Dr. Zainal Arif bin Burhanuddin
Project Supervisor
May 2012
i
CERTIFICATION OF ORIGINALITY
This is to certify that I am responsible for the work submitted in this project, that the
original work is my own except as specified in the references and acknowledgements,
and that the original work contained herein have not been undertaken or done by
unspecified sources or persons.
__________________________
ii
ABSTRACT
Dye-Sensitized Solar Cell (DSSC) has gained much interest due to its low cost of
fabrication. Many researches have been carried out to optimize its properties and
structure for maximum efficiency and fill factor. In this paper, predicting the efficiency
of a DSSC for optimization purposes using Silvaco ATLAS is outlined. Initially, DSSCs
with TiO2/dye absorption layer thickness of 6 µm, 12 µm and 18 µm were fabricated.
The absorbance of each layer was then measured using UV-Vis. From the absorbance
data, the reflection and refraction coefficients were calculated using Kramers Kronig
equations. They were later used in ATLAS to simulate the efficiency of the DSSC. In the
simulation, the DSSC was modeled as two organic sandwiched regions which consist of
TiO2 coated dye and electrolyte. AM 1.5 spectra is used as a light source. For each single
photon from the light source, a singlet is assumed to be created. Singlet equation is then
solved accounting for Langevin recombination and singlet exciton dissociation. The
outcome from ATLAS shows that better efficiency can be achieved using
18µm thick TiO2/dye absorption layer. The outcome is validated by measuring the actual
efficiency of the fabricated DSSC. Close agreement between the predicted and measured
efficiency suggest that ATLAS can be used as a tool to optimize the DSSC.
iii
ACKNOWLEDGEMENTS
First and foremost, I would like to express my appreciation to my project supervisor, Dr.
Zainal Arif bin Burhanudin for his valuable input, guidance and effort throughout this
project. It really help me a lot.
Next, I would like to thank the Solar Lab members for the supports that they provided
during the completion of the project. Thanks to all students, lecturers and also
postgraduate students from UTP especially Mr. Adel Eskandar bin Samsudin who had
provided untiring guidance and help throughout the period of the project.
Finally, I would like to apologize if any party was unintentionally excluded from being
mentioned above. Thus, I would like to thank everyone that involved in making this
project a success. Thank you very much.
iv
TABLE OF CONTENTS
v
2.3.1 Absorption coefficient........................................................................ 9
vi
REFERENCES ................................................................................................................. 42
APPENDICES .................................................................................................................. 36
LIST OF TABLES
LIST OF FIGURES
vii
Figure 11: Cold Press Process .......................................................................................... 19
Figure 21: Effect of TiO2 thickness on the short circuit current, Isc of DSSC ................. 29
Figure 22: Effect of TiO2 thickness on the open circuit voltage, Voc of DSSC .............. 29
Figure 23: Effect of TiO2 thickness on the maximum power, Pmax of DSSC ................ 30
Figure 24: Effect of TiO2 thickness on the maximum voltage, Vmax of DSSC ............. 30
Figure 25: Effect of TiO2 thickness on the Fill Factor, FF of DSSC ............................... 31
viii
CHAPTER 1
INTRODUCTION
Nowadays, silicon based solar cells is dominating the solar cell market. However, the
main issue is the production cost that is even higher than the output power produced
by it. Dye sensitized solar cell (DSSC) is anew technology of solar cell which can be
made from various material. But, this type of solar cell is having issue regarding its
efficiency level. Thus, many researches have been done in order to increase the
performance of this DSSC.
In pursuing higher efficiency of DSC, numerous researches have been done to study
and understand the mechanism of DSC. Up to now, efficiency for DSSC still lower
than Silicon based solar cell. So, instead of using experimental method in analyzing
the DSSC performance, there is a need to identify or develop software that can
analyze the efficiency of DSSC.
1
Scope of study:
1.4 Relevancy
This paper wants to verify the ATLAS ability in simulating the DSSC performance. if
this software can accurately analyze the DSSC performance as close as the experimental
value, thus this software proven reliable.
1.5 Feasibility
The feasibility of this project is to complete the project within the allocated time frame,
while maintaining the consistency of this project.
During the first semester (FYP I), the scope and task that will be covered are:
2
CHAPTER 2
LITERATURE REVIEW
Nowadays, Silicon based solar cell is widely used with approximately 25% in
efficiency. But, the weaknesses is the high production cost which is costly than the
source of energy from fossil and also difficult to fabricate. In 1991, DSSC had been
invented by Michael Grätzel and Brian O'Regan at the École Polytechnique Fédérale de
Lausanne. [11-12] DSSC is an electrochemical device that generates electricity from the
sunlight using the dye-molecules which are light-absorbing adsorbed on semiconductor
particles.[1] DSC contains mesoporous nanocrystalline semiconductors like TiO2, ZnO
and SnO2 as photoanodes anchored with dye molecules which serve as light harvesters
fabricated on Transparent Conducting oxide (TCO), a platinum (Pt) counter electrode
and an electrolyte solution with dissolved iodide ion/ tri- iodide ion redox couples
between electrodes. [6] Dye sensitizers function as light absorber and then exploit the
light energy to induce vectorial electron transfer reaction. [6]
Where FF is fill factor, is short circuit current, is open circuit voltage and is
incident power. This equation shows that in order to raise the conversion efficiency,
these four parameters are necessary to be improved. In DSSCs, it series resistance consist
of three resistance elements which are sheet resistance of TCO, the resistance of ionic
diffusion in the electrolyte and the resistance at the interface between the counter
electrode (CE) and the electrolyte. [6]
Due to low production cost, DSSCs are promising for more cost effective than
conventional solar cells. This low production cost is because of cheap printable materials
4
and simple manufacturing techniques. [1] However, less efficiency of power conversion
is the main issue that has been under research for past 10 years besides its shorter
operating lifetime. For now, some reported that DSSC efficiency currently improved to
11.5% and the other reported it about 10% - 10.8%. [6-7, 9] However, insufficient
understanding about DSC mechanism behavior caused us to rarely achieved efficiency of
over 10%. [6] Present focus is to find all factors that can lead to higher energy
conversion, lower cost and also longer operating lifetime. [2]
5
Figure 1: DSSC structure [1]
There are many photoelectrodes made from such material as Si, GaAs, and CdS.
However, oxide semiconductor materials have few advantages compare to the other
types. First, TiO2 a good chemical stability when exposed under visible irradiation in
solution. Second, they are cheap and nontoxic. Third, TiO2 thin film can be prepared by
only using a very simple process. TiO2 film normally contains large TiO2 nanoparticles
which is around 250-300 nm. This TiO2 film can be prepared by a very simple process
where TiO2 colloidal solution or paste is coated on the TCO substrate and then sintered at
450 to 500OC
6
2.2.1.2 Redox electrolyte
Most common electrolyte used in DSSC is I-/ redox ions which functions as
the electron mediator between the TiO2 photoelectrode and the counter electrode.
Reduction of the dye cations with I- ion will form the tri-iodide ion which is which
then will be re-reduce to I- ions at the counter electrode. For the counter electrode to be
able to do this task, it must have high electrocatalytic activity. Thus, platinum, Pt which
which coated on TCO substrate will usually use as the counter electrode.
2.2.2 Operation
Basically, when the dye absorbs photon from the sunlight, electron will excited
from low energy state to a higher energy state of the molecule (S0 to S1). [1, 10] Next,
electron will be injected to the TiO2 band and will be transported to TCO layer. [1, 3]
electron donation from the electrolyte such as iodide/tri iodide couple will restore the
original state of the dye subsequently. [8] Electron captured from the redox electrolyte
will regenerated the oxidized dye. The injected electron will travel by diffusion in the
TiO2 film. Electron mobility in TiO2 is about 40 to 1 cm2 s-1 V-1. [3] When it finds its way
to substrate contact, it will be released to the external electrical circuit. Illumination that
corresponds to the difference between the Fermi level of the electron solid and the redox
potential of the electrolyte will generate the voltage. [8] Then, by electrolyte reduction
reaction at the counter electrode, the electron is returned to the cell. The ionic transport
of the redox pair in the electrolyte will complete the electrical circuit and thus generated
electric power without suffering any permanent chemical transformation. [1, 8]
Operating cycle can be summarized as per equation below taken from [1]:
For photoelectrode:
7
( ) injection of electron [1]
Total reactions:
( ) ( ) ( ) photoelectrode [1]
( ) ( ) cell [1]
8
2.3 Optical properties
Optical properties of the substances define as the changes that the light undergoes when
interacting with a particular substance. There are many optical properties if we were to
consider, but the most well know are reflection, refraction, transmission and absorption.
Those optical properties are associated with most important optical properties like
absorption coefficient, extinction coefficient and also refractive index.
Absorption and absorption coefficient is closely related since absorption coefficient can
be obtained using the absorption value. The definition of absorption coefficient is really
important because some use the natural log while the other some use log to the base 10.
When illumination of wavelength and intensity impinges on a material thickness which
denotes as d, some part light will be reflected while the other part will be emerged to the
other side of the material at d. These I, R and T are related by the expression below:
( ) ( ) ( ) ( ) (1)
I(λ), T(λ), R(λ) and A(λ) are intensity, transmittance, reflectance and absorption taking
place for this wavelength. Next, by applying the Beer-lambert law, the relationship
between T and I-R become
( )
( ) [ ][ ] (2)
( )
( ) [ ][ ] (3)
( ) ( )
( ) ( )
(4)
9
( )
( ) [( ] (5)
) ( )
( )
( ) (6)
For this project, the author used equation (6) to obtain the absorption coefficient value.
(7)
(8)
Equation (8) had been used throughput this project to obtain the extinction coefficient or
k value.
Refractive index, n of an optical medium is actually the ratio of the velocity of the light
which is c in vacuum to its velocity, v in the medium, n= c/v. complex refractive index,
denotes by N is consists of n as the real part and k as the imaginary part. In short, N can
be expressed as below:
N= n – iK (9)
In order to obtain the n, Kramers-Kronig relation can be used since the extinction
coefficient which is the imaginary part can be calculated beforehand. This requires the
value of frequency dependence of either the real or imaginary part over a wide frequency
range.
10
The kramers-kronig relation is given by [16],
( )
( ) ∫
( )
( ) ∫
Where the integration variable and P represents the Cauchy principal value of the
integral and the singularity at is avoided. and is the real part and the
imaginary part. Since the Kramers- Kronig involved with complicated measurement, the
calculation of the real part will be done using MATLAB.
11
CHAPTER 3
METHODOLOGY
Problem Statement
Objectives
Preliminary Research
Literature Review
Do experiment
Result evaluation
Report preparation
12
3.2 Assumption
For this project, an assumption has been made to simplify the experiment and focus
to only one factor which is TCO layer thickness variation which is setting n and k value
to be constant where n and k is electron concentration and Boltzmann’s constant.
For the experiment procedure, the author needs to get the absorbance data for
every thickness. The author uses three different thicknesses which are 6 m, 12 m and
18 m. These absorbance data will be obtain through UV-VIS experiment which at the
same time the author will also get the transmittance value. Next, using this absorbance
value, the author will obtain the absorption coefficient data and also extinction
coefficient data, k. after that, by using Kramers - Kronig method, value of refraction
index will be obtain. All these data will then be inserted into the coding for the
ATHENA. For every thickness, these software will then evaluate the performance the
DSSC for every thickness.
3.5 Equipment
For now, these are the expected equipment needed for this experiment:
1. Dye Sensitized Solar Cell
13
3.6 Atlas software
ATLAS is a simulator that able to simulate the electrical, optical and thermal behavior of
the semiconductor devices based on virtual fabrication of its physical structure. It is able
to provide the I-V characteristic, photogeneration mapping and spectral response. For
silicon based technologies, S-Pisces is used together with the ATLAS as a device
simulator that incorporates both drift-diffusion and energy balance transport equations
[17]. In addition, Luminous is also used with the ATLAS as an advanced device model
that able to model light absorption and photogeneration in planar and non-planar
semiconductor devices.
14
3.8 Obtaining Complex Refractive Index using UV-Vis Absorption
Spectroscopy
calculate
extinction refractive
Absorbance, A absorption
coefficent, k index, n
coefficient, α
Absorbance and transmittance are obtained using UV-VIS method. Firstly, the
sample and the reference have been put in the chamber. Next, wavelength range
of the light has been set from 300nm until 800nm. Then, the chamber is closed
and measurement is started. The author calculates absorbance value for every
wavelength which ranges from 200nm until 800nm. Absorption coefficient can
be obtained using absorbance data. Next, these sets of absorption coefficient
values are used to calculate the extinction coefficient. These two values are then
used to calculate the refractive index, n.
For this part, the author has the collaboration with the postgraduate student to
accomplish this experiment. This experiment has two parts:
1. Preparation
Preparing the things need to be use in the experiment such as TCO glass,
TiO2 paste, and dye.
15
2. Screen printing
Through this method, the TiO2 paste had been printed on the TCO surface.
The author made three different thicknesses which are 6um, 12um and 18um.
One pass will make a 6 um of thickness. So eventually, we need to make 2
passes for 12 um and three passes for 18um.
This TiO2 paste consists of nanoparticles only.
16
3. Dry the printed TCO glass
For this process, the TCO glass will be dried out before proceed to the next
step. This printed TCO will be dry using belt furnace with the temperature
inside the furnace is around 550 0C.
4. Dye soaking
TCO glass will be arranged in a special box. After that, dye solution will be
poured into the box until full. Next, the box will be closed for 15 hours. This
is to ensure that the soaking process is fully complete. However, this soaking
process is varies with the type of dye used. This soaking process may takes
longer up to 48hours.
17
Figure 8: arranging the dye into the box
After 15 hours has been passed, the glass will be taken out from the box.
Next, it will go through the next process which is sandwiching the dye by
another TCO glass that has a small hole in the center. These two glasses will
be paste together using a plastic gasket that has been placed in between these
18
two glasses. Next, through the heat press process, these two glasses will be
stick together. After that, a cold press is used to cool down the heated glass.
Next is the electrolyte filling process which will be injected through the small
hole using the pump and vacuum dessicator. Last part for this step is to seal
the small hole by using the aluminum foil.
20
3.10 Simulation of solar cell
The Dye-Sensitized Solar Cell consists of two regions which are the bulk
heterojunction region and the electrolyte region. The bulk heterojunction region is TiO2
particles interconnected with dye which is light absorbing material with three different
thicknesses of film which are 6µm, 12µm and 18 µm. Electrolyte region consists of
redox solution with the thickness of 25 µm.
21
Figure 15: DSSC structure for 12µm in ATLAS
23
3.11 Obtaining IV Characteristics of the Dye-Sensitized Solar Cell
ATLAS is also used to simulate the IV characteristic of the device. Before that,
some set up need to be done. Source beam is located in the middle of the solar cell and
2µm from the cell. Angle of the beam is 90o which is normal to the cell. Air Mass 1.5
(AM 1.5) is used to test the solar cell which is the standard spectrum at the earth surface.
Parameters are listed in the table 2.
Parameter Value
Location of the beam source (x,y) (0.5,-2.0)
Angle of beam 90.0o
Solar spectrum AM 1.5
Table 2: Beam parameter set up in ATLAS
24
Chapter 4
TiO2 thickness is varies with three different thicknesses which are 6µm, 12µm and 18
µm. the temperature is fixed at 700oC. The absorbance data which is obtained from the
UV-Vis method showed the figure 17.
4.500
3.500
absorbance, a.u.
6 um
2.500
12 um
18 um
1.500
0.500
From the absorbance data that the author gets, absorption coefficients are then calculated
by using the equation 6. The result of the absorption coefficient is then plotted as below:
25
absorption coefficient VS wavelength
11000
10000
absorption coefficient,α (cm-1)
9000
8000
7000
6000
6um
5000
12um
4000
3000 18um
2000
1000
0
350.00 450.00 550.00 650.00 750.00
wavelength (nm)
By using these absorption data above, the author then calculate the extinction coefficient
using equation 8. Lastly, refractive index is determined using the kramers-kronig
equation calculated using MATLAB.
26
extinction coefficient, k vs wavelength (nm)
0.03
0.02
0.015 6 um
12 um
0.01
18 um
0.005
0
350.00 450.00 550.00 650.00 750.00
wavelength (nm)
0.006
0.004
0.002
refractive index, n
0 18 um
350.00 450.00 550.00 650.00 750.00
-0.002 12 um
18 um
-0.004
-0.006
-0.008
-0.01
wavelength (nm)
27
4.2 Dye-Sensitized Solar Cell parameter variation for simulation and experimental
The Dye-Sensitized Solar Cell was used to simulate the effect of TiO2 layer by
varying the thickness of the TiO2 layer. Simulation results shows that TiO2 layer with the
thickness of 18µm has the best efficiency with the value 4.27%.
SIMULATION
Experiment also show that TiO2 layer with the thickness of 18µm has the best efficiency
with the value 3.10%.
EXPERIMENT
The maximum short circuit current, Isc and open circuit voltage, Voc affected the Fill
Factor as shown in figure 25. For experiment, Fill Factor is the lowest for the thickness
of 18µm which differ from simulation which says that 12µm thickness has the lowest Fill
28
Factor. This difference happened because Fill Factor is the ratio between the maximum
power to the product of open circuit voltage and short circuit current. Short circuit
current, Isc variation for the experiment is not much compared to the simulation. Thus, it
gives the different result for the lowest Fill Factor.
Isc Vs Thickness
30
Short Circuit current,Isc (mA)
20
simulation
10 experiment
0
6 12 18
Thickness (um)
Figure 21: Effect of TiO2 thickness on the short circuit current, Isc of the DSSC
Voc
0.8
0.7
Open Circuit Voltage,Voc (V)
0.6
0.5
0.4
experiment
0.3
simulation
0.2
0.1
0.0
6 12 18
Thickness (um)
Figure 22: Effect of TiO2 thickness on the open circuit voltage, Voc of the DSSC
29
Pmax Vs Thickness
5
Figure 23: Effect of TiO2 thickness on the maximum power, Pmax of the DSSC
Vm Vs Thickness
0.5
Voltage at maximum power
0.4
pint,Vmpp (V)
0.3
0.2 experiment
simulation
0.1
0.0
6.0 12.0 18.0
Thickness (um)
Figure 24: Effect of TiO2 thickness on the maximum voltage, Vm the DSSC
30
FF Vs Thickness
0.6
0.5
0.4
Fill Factor,FF
0.3
experiment
0.2
simulation
0.1
0.0
6 12 18
Thickness (um)
Figure 25: Effect of TiO2 thickness on the Fill Factor, FF of the DSSC
Efficiency Vs Thickness
5
4
4
Efficiency (%)
3
3
2 simulation
2 experiment
1
1
0
6 12 18
Thickness (um)
31
4.2.1 IV characteristics of the Dye-Sensitized Solar Cell
By plotting the current and voltage, there are significant differents for every thickness
with 18µm has the highest curve compared to the other two thicknesses.
Current Vs Voltage
16
14
12
Current (mA)
10
8 6um
6 12um
4 18um
2
0
0 0.2 0.4 0.6 0.8
Voltage (V)
Power Vs Voltage
5
4.5
4
3.5
3
Power (mW)
2.5 12um
2 6um
1.5
18um
1
0.5
0
-0.5 0 0.2 0.4 0.6 0.8 1
Voltage (V)
32
4.2.1.2 IV characteristics for experiment
Experimental result is same as the simulation result with the 18µm give the highest
curve. However, the values difference between three thicknesses is quite close compared
to the simulation result.
Current Vs Voltage
12
9
Current (mA)
6 6um
12um
3 18um
0
0 0.2 0.4 0.6 0.8 1
Voltage (V)
Power Vs Voltage
3.5
2.5
Power (mA)
2
6um
1.5
12um
1 18um
0.5
0
0 0.2 0.4 0.6 0.8 1
Voltage (V)
33
Chapter 5
Firstly, the student had performed the calculation of the refractive index
used in Dye-Sensitized Solar Cell. Calculated refractive index has shown the
effect of varying the thickness of TiO2 layer. Thicker layer of TiO2 resulted to
higher value of refractive index.
This has been done at solar lab and the data obtained will be visualized in
the form of graph. In order to obtain accurate data, these experiments need to be
conducted in a very careful way since DSSC is very small. A deep understanding
on DSSC mechanism is a must since it will affect the experimental result later on.
From the experimental result, 18um has the best efficiency compare to
6um and 12um. Simulation result also gives the same result which is 18µm gives
the best efficiency compared to 6µm and 12µm. The author first assumption is
that thicker layer of TiO2 will give the opportunity for the dye to absorb more
sunlight and thus increase the dye-sensitized solar cell (DSSC) efficiency.
Simulation and experimental result prove this.
However, even though both simulation and experimental result point out
the same thickness for the best efficiency, the efficiency value between
simulation and experimental is quite big. Simulation shows that thickness of
18µm gives efficiency of 4.27% while the experimental gives 3.10%. It gives
37.6% different between these two results. This large difference may happen due
34
to error made during fabrication of Dye-Sensitized Solar Cell in the lab and also
some assumptions made during simulation process. In order to get more accurate
result, repetitive experiment with at least three samples for every thickness is
suggested.
Lastly, both experimental and simulation are able to show the pattern of
the effects of the thickness variations in TiO2 layer.
35
APPENDICES
Start
Yes
Change
thickness?
No
End
36
APPENDIX B: Gantt chart
No Detail 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 Selection of Project Title:
2 Preliminary Research Work: Research related to topic
37
APPENDIX C: MATLAB CODE [16]
function imchi=kkimbook(omega,rechi,alpha)
%The program inputs are 1) omega, vector of the frequency
%(or energy) components, 2) imchi, vector of the imaginary
%part of the susceptibility under examination, and 3) alpha,
%the value of the moment considered. The two vectors
%1) and 2) must have the same length.
%The output is the estimate of the real part as obtained
%with K-K relations.
%In order to use this program, save the whole text contained
%in this section in a file and name it kkrebook.m
load omega.csv
omega=f;
%omega(:,1)=[]; %eliminate first column to get omega vector
if size(omega,1)>size(omega,2);
omega=omega';
end;
load kenam.csv
imchi=kenam;
%imchi(:,1)=[];
if size(imchi,1)>size(imchi,2);
imchi=imchi';
end;
alpha=0;
j=1;
omega=j*omega;
beta1=0;
for k=2:g;
b(1)=beta1+imchi(k)*omega(k)^(2*alpha+1)/....
(omega(k)^2-omega(1)^2);
beta1=b(1);
end;
rechi(1)=2/pi*deltaomega*b(1)*omega(1)^(-2*alpha);
%First element of the output: the principal part integration
%is computed by excluding the first element of the input
38
j=g;
alpha1=0;
for k=1:g-1;
a(g)=alpha1+imchi(k)*omega(k)^(2*alpha+1)/...
(omega(k)^2-omega(g)^2);
alpha1=a(g);
end;
rechi(g)=2/pi*deltaomega*a(g)*omega(g)^(-2*alpha);
%Last element of the output: the principal part integration
%is computed by excluding the last element of the input
for j=2:g-1;
%Loop on the inner components of the output vector.
alpha1=0;
beta1=0;
for k=1:j-1;
a(j)=alpha1+imchi(k)*omega(k)^(2*alpha+1)/...
(omega(k)^2-omega(j)^2);
alpha1=a(j);
end;
for k=j+1:g;
b(j)=beta1+imchi(k)*omega(k)^(2*alpha+1)/...
(omega(k)^2-omega(j)^2);
beta1=b(j);
end;
rechi(j)=2/pi*deltaomega*(a(j)+b(j))*omega(j)^(-2*alpha);
end;
%Last element of the output: the principal part integration
%is computed by excluding the last element of the input
39
APPENDIX D: ATLAS CODING
40
log off
save outf=check2.str
extract init infile="light_current.log"
extract name="Isc" y.val from curve(v."cathode", (i."cathode"))
where x.val=0
extract name="Voc" x.val from curve(v."cathode", (i."cathode"))
where y.val=0
extract name="Jsc (mA/cm2)" abs($"Isc")*1e-1
extract name="Power" curve(v."cathode", (v."cathode" *
i."cathode"
*(-1))) \
outf="P.dat"
extract name="Pmax" max(curve(v."cathode",
(v."cathode"*i."cathode"*(-1))))
extract name="V_Pmax" x.val from
curve(v."cathode",(v."cathode"*i."cathode")) \
where y.val=(-1)*$"Pmax"
extract name="Fill Factor" ($"Pmax"/(abs($"Isc")*$"Voc"))
extract name="intens" max(probe."inten")
extract name="Eff" (($Pmax/($"intens"*1e4))*100)
tonyplot dark_current.log -overlay light_current.log
tonyplot P.dat
tonyplot check2.str
quit
41
REFERENCES
1. Janne Halme, P. V. (Sep 2010). Device Physics of Dye Solar Cells. Advanced Materials, E210
- E234.
2. Kawakita, J. (dec 2009). Trends of Research and Development of Dye-Sensitized Solar Cells.
3. Yanagida, S. (May 2003). Dye Sensitized Solar Cells: Management of Photoelectrons. 3rd
World Conference on Phorovolroic Energv Conversion. Osaka, Japan.
4. H. Hosonoa, H. O. (2002). Vacuum. Frontier of transparent conductive oxide thin films, 419-
425.
5. C. Guillén, J. H. (2011). Thin Solid Films. TCO/metal/TCO structures for energy and flexible
electronics, 1-17.
6. Yasuo CHIBA, A. I. (2006). Dye-Sensitized Solar Cells with Conversion Efficiency of 11.1%.
Japanese Journal of Applied Physics, L638-L640.
7. Wei, D. (2010). Dye Sensitized Solar Cells. International Journal of, 1103-1113. Retrieved
from www.mdpi.com/journal/ijms
9. Liyuan Han, A. F. (n.d.). High Efficiency of Dye-Sensitized Solar Cell and Module.
10. Halme, J. (Feb 2002). Dye-sensitized nanostructured and organic photovoltaic cells:technical
review and preliminary tests.
12. Komal Magsi, E. M. (n.d.). Calculating the Efficiency of Dye Sensitized Solar Cells with
Industry Based Silicon Solar Cells
13. http://solar.calfinder.com/blog/solar-information/solar-genealogy-on-three-generations-of-
solar-cells/
14. http://www.solar-facts-and-advice.com/solar-cells.html
42
15. Gourdin G., Introduction to Green Chemistry(2007)
16. Lucarini V., Peiponen K., Kramers-Kronig Relations in Optical Material Research(2005)
43
44