Mdp36 the End

Nanotechnologies for Photovoltaic Solar Electricity Generation

Design Group 36

Michael King, MEYen-Cheng Lin, MESerdar Marcan, CSArpit Mehta, ME

Faculty Advisor: Alok Rastogi

May 8th, 2009

Submitted in fulfillment of the requirements of ME494/EECE488 in the Spring Semester of 2009.

Thomas J. Watson School of Engineering and Applied ScienceState University of New York at Binghamton

Abstract

The need for inexpensive, clean renewable energy is a serious problem requiring complex

solutions. A way to provide the necessary energy in a less expensive solar cell is demonstrated in

the dye-sensitized solar cell (DSSC). Provided is a detailed method of constructing these cells

complete with a bill of materials and explanation of the use of each material. DSSC currently

cannot compare with the traditional silicon based photovoltaic solar cells, however with an ease

of manufacturability at a fraction of the cost makes this product an interesting alternative for

developing countries. Research was an essential portion of the project, providing techniques for

developing the titanium dioxide deposition and dye selection. Many prototypes were created in

an effort to perfect the mechanical techniques used when fabricating these cells by hand.

Advanced testing was crucial to provide the feedback necessary to hone those skills and enhance

the efficiency of the cells.

Table of Contents

1. Introduction 1

2. Theory 3

3. Prototyping 53.1 Bill of Materials 53.2 Purpose of Each Material 53.3 procedure for Dye-Sensitized Solar Cell 6

4. Research 114.1 Discovering laboratory techniques 114.2 Titanium Dioxide Deposition & Sintering 114.3 Small Glass to Larger Glass 124.4 Natural Dye to Synthetic Dye 134.5 Natural Dye in DSSC 14

5. Analysis 145.1 Synthetic Dyes 195.2 Testing Voltage versus Current 195.3 Calculating Efficiency 22

6. Conclusion 24

- Appendix A: Requirement Matrix A-1

- Appendix B: Block Diagram A-3

- Appendix C: Budget A-4

- Appendix D: Schedule A-5

- Appendix E: Laboratory Pictures A-6

List of Figures

Figure Number

Name Page Number

1 Electron Flow in Titanium Dioxide Dye-Sensitized Solar Cell 2 Completed Sintering process 2 inches by 2 inches plate 3 Completed Sintering process 1 cm by 1 cm plate

4 Soaking dye with sintered TiO2 Cell 3

5 Tin Oxide coating glass with liquid graphite layer 6 Assemble the two glass plates 7 Clamp the plates together 8 Cell, Electrolyte liquid and needle 9 Insert the electrolyte solution.

10 Sealing off each edge to protect the electrolyte solution from the atmosphere. 11 Nitrogen tube furnace

12 Current (micro-Amperes) versus Time (seconds) for Blackberry Solar Cell with Pencil Backing

13 Voltage (mV) versus Time (seconds) for Blackberry Solar Cell with Pencil Backing

14 Current (micro-Amperes) versus Time (seconds) for Blackberry Solar Cell with Graphite Backing

15 Voltage (mV) versus Time (seconds) for Blackberry Solar Cell with Graphite Backing

16 Current (micro-Amperes) versus Time (seconds) for Pomegranate Solar Cell with Graphite Backing

17 Current (micro-Amperes) versus Time (seconds) for Beet Solar Cel l with Graphite Backing

18 Voltage (mV) versus Time (seconds) for Pomegranate Solar Cell with Graphite Backing

19Voltage (mV) versus Time (seconds) for Beet Solar Cell with Graphite Backing

20 Current (Amperes) versus Voltage (Volts) for Blackberry Solar Cell Without Illumination

21 Current (Amperes) versus Voltage (Volts) for Blackberry Solar Cell Under Tungsten-Halogen Lamp

22 Current (Amperes) versus Voltage (Volts) for Blackberry Solar Cell Under Illumination; Focus on Maximum Power Produced As Shown By Rectangle

List of Tables

Table Number

Table Name Page Number

1 Current and Voltages of Synthetic Dyes

1. Introduction

Energy is one of the greatest problems facing current and future generations. Population growth

combined with development in foreign countries has caused the world demand for energy to

increase greatly. According to the Energy Information Administration [1], world demand for

energy will increase by fifty percent from 2005 to 2030, from 462 to 695 quadrillion (1015)

British Thermal Units (BTUs). Much of this growth comes from developing countries such as

India and China.

Unfortunately, the majority of this energy will come from unclean sources such as gas and coal.

As described by the Kuznets curve, when a country initially develops it will take on

environmentally unsound practices while it grows. Countries that have already developed can

afford to enact laws protecting the environment and restrict the amount of pollution coming from

energy production. Natural gas and coal are both the fastest growing forms of energy production

both of which rely on tapping a limited natural resource and pollute the environment when

burned.

The sun provides a potentially unlimited source of energy, at least for the next five billion years.

Humans consume 4.1 x 1020 Joules of electricity per year, while the sun provides 4.3 x 1020

Joules of energy to the Earth in one hour [2]. Not all this energy is harvestable, since most fall on

areas where it cannot be captured and no device will have 100% efficiency, but it shows that the

sun provides an incredible amount of energy. Only a small portion of this energy need be

captured to power humanity’s needs.

Solar electricity’s primary advantage is that it is based on an unlimited, clean resource. Unlike

the energy utilized from coal, solar energy does not output any harmful byproducts when it is

utilized. It does not emit carbon dioxide or greenhouse gases to produce electricity. It is also an

unlimited resource since the sun will continue to provide light to the earth for the next five

billion years. Unlike other resources used for energy, such as petroleum, the sun also is not

controlled by a nation or organization. The sun is a completely independent resource.

The main impediment to the large scale use of solar electricity is economics. Traditional solar

cells, made of silicon, are not cost effective. High purity silicon is the foundation of most

computers and processor products and is naturally a very scarce resource, resulting in a more

costly product. Silicon based cells do not perform well in diffused light, meaning that installation

must be on a roof or an area with an unobstructed view of the sun. Commercial silicon cells are

also quite large. This may become a problem for installation on private residences where the

overall load of the cell combined with the housing component might be too large for the

structural capabilities of the roof. This comes at an added cost to the consumer. The cost of solar

cells must be decreased by at least 25 times in order to compete with fossil fuels. Novel

approaches to solar technology must be considered in order to create a cell that is suitable for

wide-scale use.

The major technical advantage with such nanocrystalline solar cells is the fact that the surface

area of grain boundaries is very high, significantly increasing the density of recombination. The

nanostructure of the titanium dioxide (TiO2) layers acts as a semiconductor “sponge”, absorbing

hundreds of dye monolayers to receive photons from light and release electrons to the titanium

dioxide [3].

These structures also have potential for high absorption due to a higher density at the band edge.

Nano-sized surfaces will have a very close contact area, significantly decreasing the band gap

and increasing efficiency. The larger the band gap the more energy is required to move the

electrons from one end of the junction to the other. Nanowires provide a more direct path for

charge transport to the contacts; whereas dendritic nanowire structures also allow for enhance

light harvesting [4].

The efficiency is increased with decreasing titanium particle size and increasing titanium layer

thickness. Smaller particle size creates larger surface areas for the sensitizing dye to adhere to,

while allowing for greater sun penetration depth. The thicker layers create more electron flow

over the exposed surface area, theoretically creating more current.

Theory

In traditional solar cells silicon performs two primary functions. First, the silicon must accept the

photon from light. Then it must create an electric field to separate the charge particles and create

the flow of electricity. Dye-sensitized solar cells (DSSC) divide the functions into separate

components of the solar cell, as shown in Figure 1.

Figure 1: Electron Flow in Titanium Dioxide Dye-Sensitized Solar Cell

The components of a DSSC are sandwiched between two conductive glass plates. The glass

plates are coated with a layer of tin oxide to allow the flow of electricity across the glass. As

shown in the figure, titanium dioxide is deposited on the top glass plate layer. The

nanocrystalline titanium dioxide layer is highly porous with large a surface area allowing

electrons to freely flow through its surface to the conductive glass layer. Titanium dioxide takes

on the function of charge separation seen in traditional silicon solar cells by allowing the

negative charges to flow through its surface.

As shown in Figure 1, a dye is deposited onto the titanium dioxide layer. The dye takes on the

function of light absorbance since titanium dioxide cannot directly absorb photons from light.

The dye absorbs the photons, which then excites an electron in the dye. The electron is passed

through the titanium dioxide onto the conductive layer of the glass plate. This essentially creates

the anode in the circuit. However, the electron lost by the dye must be replenished. As seen in

Figure 1, the cell contains a liquid electrolyte within it. This is an iodide/triiodide solution.

Iodide gives an electron to the dye, which converts it to triiodide.

Triiodide must be converted back to iodide. The lower layer of conductive plate is coated with a

backing on the conductive side, most commonly consisting of either platinum or carbon. The

backing layer makes up the counter electrode in the circuit and returns an electron to the

electrolyte solution. This makes up the cathode in the dye sensitized solar cell and completes the

circuit. Electricity is the flow of electrons. This cell allows electrons to flow from the dye and

titanium dioxide to power an object, the electrons are then returned through the carbon backing

and electrolyte solution.

The entire process is similar to artificial photosynthesis. A synthetic or organic dye takes the

place of chlorophyll in plants, though actual chlorophyll can be used. Titanium dioxide replaces

carbon dioxide as the electron acceptor in photosynthesis. The electrolyte solution and carbon

backing replaces water and oxygen as the electron donor. The process is regenerative because no

one material is entirely used up by the process [5].

The reactions found in the solar cell are:

Dye + Photon Excited DyeExcited Dye + Titanium Dioxide Free Electron in Titanium Dioxide + Oxidized Dye

Oxidized Dye + 3/2 I- Dye + ½ I3-

½ I3- + e- (from carbon backing) 3/2 I-

3. Prototyping

3.1 Bill of Materials

EthanolNitric/Acetic Acid Conductive glass plates with the Tin oxide (SnO2) coatingNanocrystalline Titanium dioxide (TiO2) Degussa P25 powder Various Dyes- blackberry, beetroot juice or pomegranate juice, Gallium(III),

Coumarin 153, Styrl 7 and B3 (N4) Ruthenium dyeLiquid graphite – Water and Isopropyl based Ethylene Glycol Potassium iodide Potassium iodineIsopropanol alcoholBunsen burnerMultimeterConvection oven (or furnace)Binder clipsGlass rodMortar PestleTapeAlligator clipSolar illuminationHardman’s EPOXY resin based two part sealant Voltage current providerComputing efficiency software program

3.2 Purpose of Each Material

Titanium dioxide (Nanocrystalline TiO2 Degussa P25 powder) Covered with sensitized dye Dye donates electrons to TiO2 TiO2 passes electrons to silicon dioxide and then to the external circuit

Conductive glass plates (Tin oxide or SnO2 coating) Two glass plates hold the solar cell, top glass having layer of TiO2 and

graphite layer is on the bottom glass Glass allows light (photons) to pass through to dye Conductive coating of Tin oxide (SnO2) on glass receives electrons from

TiO2, passes it to electrical circuit

Sensitized Dye Coats TiO2

Receives photons and releases electrons through TiO2 to conductive glass

Absorbs light radiation

Electrolyte (Potassium iodide and iodine (Iodide/electrolyte solution)) Surrounds TiO2 in solar cell Creates voltage potential between itself and TiO2 Allows flow of electrons by voltage potential Restores electrons to dye to maintain circuit

Backing (which is graphite or platinum layer, is on the bottom glass) Completes electric circuit by providing counter-electrode Provides electron to triiodide to convert it back to iodide

Sealant The electrolyte solution is very sensitive to light and the water or even

moisture from the air, therefore seal off the cell to prevent cell interaction with air.

3.3 Procedure for Dye-Sensitized Solar Cell [6]1. Lab coat and gloves are advised when working in chemistry labs.

2. Identify the conducting side with the tin oxide coated glass piece by using a

multimeter to measure resistance. The conductive side will have a resistance

of 20-30 ohms with 2 inches by 2 inches area of glass.

3. Tape the glass with 3~4 mm wide tape on the one side edge at conducting side

up for electrical contact, then wipe off any contaminants or oils by using

tissue wet with Acetone.

4. Dilute Acetic acid to a pH level of 3, slowly pour the acid into the water,

down the side of the container - never pour the water into the acid.

5. Grind about 1 gram of nanocrystalline titanium dioxide (TiO2) Degussa P25

powder in a mortar and pestle while gradually adding 1.66 ml diluted Acetic

acid until obtain a colloidal suspension with a smooth consistency, somewhat

like latex paint.

6. Replace some of the TiO2 suspension (paint) and quickly apply on the

conductive side of glass plate evenly with a glass rod. Proper spacer thickness

is recommended to be 7~10 micrometers of TiO2. (The TiO2 layer may

propagate large cracks and peel off easily if the TiO2 layer is too thick)

Quickly move the glass plate to pre-heated convection oven with 100~150

degree Celsius to dry.

7. Carefully remove the tape without scratching the TiO2 layer.

8. The sintering process in the convection oven or furnace occurs at 450 degrees

Fahrenheit for approximately thirty minutes. It has completed once the surface

turns to a light green color while heating. The titanium dioxide changes back

to white once the sample has returned to a lower temperature, as shown in the

following Figures 2 and 3.

Figure 2. Completed sintering process for 2 inches by 2 inches plate

Figure 3. Completed sintering process for 1 cm by 1 cm plate

9. After cooling off, soak with the dye of your choice until the titanium dioxide

has become completely immersed in the dye. This could take anywhere to 48

hours to a full week. A uniform color on the back of the glass side is a key

indicator to when the cell is ready, shown in Figure 4 Below.

Figure 4. - Soaking dye with sintered TiO2 Cell

10. Rinse gently with water and then with ethanol (the ethanol serves to remove

water from the porous TiO2)

11. The second tin oxide glass plate is coated with a liquid graphite layer on the

conductive side as shown in Figure 5. Allow the sample to air dry. Step 3

should be performed prior to the liquid carbon application.

Figure 5. Tin oxide coated glass with liquid graphite layer

12. Place the fully dried carbon coated glass into the convection of and bake at

150 degrees Fahrenheit for approximately twenty minutes. After removing

from the oven and cooling, test the surface of the carbon with a multimeter.

Should the resistance of the carbon still be out of the desired range of one to

three hundred ohms, place back into the oven for another ten minutes and

retest with multimeter after cooling.

13. Assemble the two glass plates with dyed TiO2 and liquid carbon layers

touching each other, but offset so that uncoated glass extends beyond the

sandwich for electrical contact as shown in Figure 6. Do not rub or slide the

plates. Clamp the plates together as shown in Figure 7.

Figure 6. Assemble the two glass plates Figure 7. Clamp the plates together

14. The electrolyte solution is made with the ratio of 10ml of Ethylene Glycol

with 0.127g of I2 and 0.83g KI. The electrolyte solution is very sensitive to

light and the water, even moisture from the air could damage the solution.

Close the bottle tightly and avoid exposure to light, as shown in Figure 8.

15. Insert few drops of electrolyte solution between the glass plates as seen in

Figure 9 and connect to a multimeter with alligator clip to each plate’s

electrical contact area to test the current and voltage produced by solar

illumination.

16. Clamp the two glass plates together and quickly seal the top, left and right

edges of the cell as illustrated in figure 8. Be sure to hold the cell upright

allowing the electrolyte to flow out of the bottom.

INSERT PIC OF 3 SIDES BEING SEALED

Figure 8. Cell, Electrolyte, liquid, and needle

INSERT PIC OF LG AREA CELL BEING INJECTED

Figure 9. Insert the electrolyte solution.

INSERT PIC OF EPOXY SEALED CELL

Figure 10. Sealing off each edge to protect the electrolyte solution from the atmosphere.

17. Once the three sealed sides has had time to properly set, approximately five to

ten minutes. Inject approximately 0.5 ml of the KI electrolyte solution.

18. The hope is to retain as much of the electrolyte solution in the cell before

being completely sealed. Once the cell no longer absorbs any of the

electrolyte, dry properly and seal with the resin based epoxy

19. Once the final edge has seal properly use a small amount of acetone and water

to clean the surfaces of the cell of any dirt or excess sealant. It may be

necessary to use a small razor to scrape and unwanted sealant away.

20. Using voltage as an input through the cell, a computer program captures

current versus voltage data. This data can be used to calculate the efficiency of

the cell.

4. Research

4.1 Discovering laboratory techniques for creating solar cell

All initial laboratory techniques were provided by Professor Alok Rastogi in the research paper

Titanium Dioxide Dye Sensitized Solar Cells [7]. No publication or author name was attached,

but the article did provide information on the newly researched cell. Adequate details into the

chemical structure of the electrolyte solution as well as the titanium dioxide layers were given.

The author suggested that similar cells have reached efficiencies of ten percent at fifty watts per

square meter (50W/m2). These are very exciting results with large scale possibilities given the

easy manufacturability of this product.

Through extensive research from peer reviewed publications, the process of creating these solar

cells was simplified. The Binghamton University Library’s electronic catalog of scientific and

technical journals was the foundation of our research in cell construction and dye selection. The

initial unnamed paper suggested using a ruthenium bipyridal dye, which is still currently used in

commercial applications. This dye is expensive and difficult to synthesize so to cut the overall

cost of this product, less expensive chemical dyes were purchased.

The co-creator of this brand of photovoltaic cells, Michael Grätzel has numerous publications

which were very important to our design process. Confirming the quoted results from the

mystery paper with information from a reputable author and source was necessary to justify our

design method. The paper Demonstrating Electron Transfer and Nanotechnology: A Natural

Dye-Sensitized Nanocrystalline Energy Converter [8] explained in much greater detail how

titanium dioxide cells have the capability of creating a current with either sophisticated

chemicals or natural dyes.

4.2 TiO2 Deposition and Sintering

The application of the titanium dioxide is of particular importance to ensure the success of

creating a highly efficient cell. The titanium dioxide is prepared in a solution of acetic acid and

water. The application of the titanium dioxide to the conductive glass can be done on a variety of

ways when in a solution form. A centrifuge can be used to ensure an even coating, but a large

amount of material is lost in the process. Spray and brush methods posed the potential of

contamination. The method chosen was a small diameter glass rod where the titanium dioxide is

rolled on in even layers. An even layer is ideal for dye absorption and to ensure that light can

transmit through the cell.

Discovering techniques for consistently developing these cells took time and many prototypes.

Mastering the application of the titanium dioxide is so important because cracks easily

propagate. This is particularly troubling because of the potential of short circuiting the cell

should the electrolyte solution make contact with tin oxide conductive backing. Only by

fabricating many titanium dioxide samples would it be possible to determine what exactly makes

a stable sample.

To combat this problem an optical microscope was used to observe which sintering method was

the most consistent with required specifications. The air flow and temperature conditions prior to

sintering were tested for a hot air gun, hot plate with steady state airflow conditions and free air

flow. Examining the size and pattern of cracks, the hot plate at 100 degrees Celsius had the best

results after several trials.

4.3 Movement from small glass plates to large glass plates

Professor Rastogi provided a small sampling of conductive glass to be used in the beginning

trials of the titanium application. Four inch square conductive glass plates were obtained to be

used in combination with synthetic dyes to produce larger current outputs. Larger glass plates

introduced new difficulties in creating a uniform layer of titanium dioxide. Another concern was

that samples could no longer fit into the nitrogen tube furnace which had since been used to

sinter all titanium dioxide shown in Figure 11. It was decided to purchase a convection oven

from a local retail store to replace the furnace.

Figure 11. Nitrogen tube furnace

Commercial convection ovens are coated inside and out with several varnishes for scratch

resistance and aesthetic look. The particulates from these varnishes contaminated the titanium

cells during the sintering process, turning the samples a red-brown color. To resolve this issue

the convection oven was left on for approximately one half hour to effectively burn off the

interior sealants. Also small aluminum protective coverings were assembled for the conductive

glass while inside the convection oven. Leaving the oven door during sintering helped to prevent

contamination. This was to promote even heating throughout the cell.

Talk about Bunsen Burners in chem lab

The overall goal has remained to create a cell which will operate with an efficiency between

three and five percent. Larger conductive glass pieces would allow for ease of creating a small

panel to be analyzed in direct and diffused light conditions.

4.4 Movement from natural dyes to synthetic dyes

It was suggested in multiple research papers that natural dyes could replace the chemical dyes at

the cost of the desired efficiency levels. This is possible because the requirements for an

effective dye are to darken the titanium, enhancing sunlight absorption and to create electrons

through a process similar to artificial photosynthesis. Pomegranate seeds, blackberries, and beet

root were used in place of the expensive chemical dyes. Natural dyes performed well but as

expected not up to the desired three to five percent because of the impurity of the solution. To

gain a larger output current from the given area, chemical dyes were necessary.

4.5 Research various types of dye used in DSSC

Michael Grätzel’s Conversion of sunlight to electric power by Nanocrystalline dye sensitized

solar cells is where an ideal dye sensitizer is described. The majority of the study in the dye is

the delicate balance between the energy producing requirements in the natural sunlight while

remaining stable in the photo-electrochemical environment. Ruthenium bipyridyl or N3 dyes

were suggested again as the most effective sensitizer quoting that “has been unmatched for eight

years by virtually hundreds of other complexes” [9]. Seeking a less costly alternative, organic

coumarin dyes showed efficiency results of approximately eight percent. Coumarin 343 is known

to excite electron movement in titanium dioxide at a much higher rate than most competitive

dyes, however the wavelength at which the coumarine absorbs the dye is lower than desired and

the range is limited.

5. Analysis

5.1 Natural Dyes

Initial solar cells were created using natural dyes from certain fruits, such as blackberries,

pomegranate, and beets. The first solar cell was created using a cell soaked in blackberry juices.

Since the liquid graphite was not obtained yet, the backing was created using graphite from a

lead pencil. The current and voltage of this cell is shown below:

Figure 12: Current (micro-Amperes) versus Time (seconds) for Blackberry Solar Cell with Pencil Backing

Figure 13: Voltage (mV) versus Time (seconds) for Blackberry Solar Cell with Pencil Backing

In both figures, the cell is initially exposed to natural sunlight. After a certain point, the sunlight

is blocked and the effects on the cell are observed. In Figure 12, the current starts off at

approximately 3 micro-amperes, but begins to decrease slowly from the initial value in sunlight.

The sunlight is blocked after 3 seconds, and the current drops precipitately. The current goes to

1.5 micro-amperes, a decrease of over half. However, there is still some current produced,

showing that the cells are able to work in diffuse light. The sunlight is blocked, but it still picks

up enough light being reflected by the surroundings to produce some current. The current for this

cell is very low compared to later natural dye solar cells.

Figure 13 shows the voltage produced by the cell. The voltage starts off at 325 millivolts and

starts to gradually increase from there. After 7 seconds, the sunlight is blocked and the voltage

begins to drop. The voltage decreases to approximately 290 mV. While the voltage does decrease

with reduced light, it does not show as large of percent decrease as the current.

Voltage measures the electrical potential between two areas. The measured voltage was as

expected, while the current was low. In a dye-sensitized solar cell, a strong voltage between the

anode and cathode implies that the cell was properly made and the components are functioning

properly. The cell is able to create a potential difference. However, the low current implies that

the electrons are not able to freely flow from the anode to cathode.

The low current is likely a result of the dye used, since fruit dyes are an impure substance.

However, greater currents were produced using fruit dyes in later experiments. In this case, the

poor current was most likely a result of the pencil backing. The pencil backing initially used

probably could not produce a full, even coating on the conductive glass plate. This would result

in a decreased current.

Solar cells were later produced using liquid graphite. Liquid graphite provided a much more

complete coating of the glass surface. Voltage and current were tested again with a blackberry

solar cell.

Figure 14: Current (micro-Amperes) versus Time (seconds) for Blackberry Solar Cell with Graphite Backing

Figure 15: Voltage (mV) versus Time (seconds) for Blackberry Solar Cell with Graphite Backing

As shown in Figure 14, the current is much larger in the solar cell with liquid graphite coating

compared to the cell with a pencil coating. The current starts off very high, at 35 micro-Amperes

and then gradually decreases to 26 micro-Amperes on its own. After approximately 15 seconds,

the sunlight is blocked. The current gradually decreases to less than 15 micro-Amperes. As in the

case of the first solar cell, blocking sunlight decreases the current by approximately 50 percent.

The cell still produces some current in blocked sunlight, showing that the cell still functions

under diffuse light.

As shown in Figure 15, the voltage is similar to that of the original cell. In this case, the voltage

is actually somewhat lower. The voltage slowly drops, even when exposed to sunlight. At

approximately 20 seconds, the light is blocked and the voltage drops slowly. It does not drop as

much as the current does in reduced light.

The experiment was repeated for different types of fruit – pomegranate and beet. At this point,

only the liquid graphite backing was used due to superior results.

Figure 16: Current (micro-Amperes) versus Time (seconds) for Pomegranate Solar Cell with Graphite Backing

Figure 17: Current (micro-Amperes) versus Time (seconds) for Beet Solar Cel l with Graphite Backing

Figure 18: Voltage (mV) versus Time (seconds) for Pomegranate Solar Cell with Graphite Backing

Figure 19: Voltage (mV) versus Time (seconds) for Beet Solar Cell with Graphite Backing

Figure 16 and 17 show current for the solar cells made from the dyes of pomegranate and

blackberry. Both the pomegranate and blackberry decrease in current from their initial values. In

the pomegranate, the sun is blocked at 18 seconds. While the solar cell continued to decrease in

current, it did not show the marked decrease in current as in the other cases. The beet solar cell

was removed from sunlight at approximately 20 seconds. It shows the noticeable decrease in

current. The current declines from 25 micro-Amperes to 15 micro-amperes.

Figure 18 and 19 show voltage for both solar cells. The pomegranate shows a strong voltage of

400 mV, while the beet shows a voltage more in line with the rest of the solar cells made. The

pomegranate voltage decreased slightly from its initial value of 418 mV. The sunlight was

removed at 13 seconds. The voltage showed increasing decline, but the voltage still remained

high. The voltage for the beet remained relatively stable at around 200 mV, though it declines

slightly through the curve. It only declined slightly when the sunlight was removed at 18

seconds.

5.2 Inittial Synthetic Dyes

The first three synthetic dyes were tested on larger conductive glass plates, with an area of 4 cm2.

The following table summarizes the result of these experiments:

Dye Current (micro-Amperes) Voltage (mV)Styryl 7 70 90Coumarin 153 150 200Gallium 40 80

Table.1 Current and Voltages of Synthetic Dyes

In each case, the current and voltages were lower than expected. The Styryl 7 dye and Gallium

dye both produced very small voltages, showing that there are deficiencies in the solar cell. The

deficiencies both seem to be attributed to the dye. The Styryl 7 did not bind to the titanium

dioxide layer. The titanium dioxide layer is usually stained the color of the dye, but the Styryl 7

color did not remain on the dye after it was removed from the solution. This dye may not be

compatible with titanium dioxide.

The Gallium dye stained the titanium dioxide a very light blue color. This dye may not be very

highly absorbent of photons or only absorbs in a narrow wavelength range, leading to the low

results from the Gallium solar cell.

Of the first three synthetic dyes, the Coumarin 153 solar cell produced the best results. It

produces a voltage similar to that of the fruit solar cells. The current is higher than those

achieved by the fruit dyes, though it is lower on a per area basis. The synthetic dyes were tested

on an area four times that of the fruit solar cells.

5.3 Ruthenium Bipyridal Dye

Coumarin 153, Gallium (III) phthalocyanine chloride and Styrl 7 were purchased, synthesized

and thoroughly tested during the fall semester and proved to be ineffective for the required

applications. After several discussions with Professor Rastogi, it was decided that the more

expensive ruthenium B3 (N4) dye would be purchased in an effort to prove that significant

power could be produced from these small area cells. A chemical manufacturer in Australia

named DyesolTM was able to provide one hundred milligrams of ruthenium bipyridyl dye and

advised to synthesize in a weak acid solution of either nitric or acetic acid. While combining the

dye with the weak acid solution it was observed that small particulates would not dissolve. An

ultrasonic water bath was easily able to break down the particulates resulting in a homogenous

liquid dye.

Logic would suggest that a more uniform, dark color would lead to an increased rate in light

absorption and thus increase the likelihood of enhanced efficiency. While experimenting with the

initial three synthetic dyes all titanium dioxide samples were placed face up in a Petri dish where

the dye was applied with a dropper until the surface was covered. This process was repeated

several times until the surface retained the color of the respective dye, shown in figure 4. Prior to

using the first batch of ruthenium bipyridal dye, Prof. Rastogi suggested submerging the cell in a

bath of the dye until all color had been removed by the nano-pores of the titanium dioxide.

Extending the amount of time which the titanium dioxide had to soak in the N3 dye along with

the immersion technique, vastly improved the dye absorption and increased efficiency was

achieved. This extension in time however greatly increased the time for one photovoltaic cell to

be completed. The difference in color associated with a longer soak time can be seen in figure

20.

Figure 20. Full bath and longer soak results in darker samples

5.4 Changing Sealants

The 832 Dow Corning multi-surface sealant was the chosen method of sealing the photovoltaic

cells from atmospheric conditions in the original cell fabrication trials. The silicone sealant

proved to be difficult to apply and would remain tacky to the touch even twenty four hours after

application. This was not sufficient for the needs described in the customer requirements matrix.

After consulting with Bob Kaub and the staff in the student design shop, a resin based two part

epoxy sealant was chosen mostly due to the fact that it totally seals within minutes. The chemical

characteristics of the sealant were researched to ensure that the sealant would not chemically

interfere with the process taking place inside of the photovoltaic cell. Switching from the silicone

to the epoxy sealant was beneficial because now construction, cleaning and testing could now

occur in a span of hours not days.

The epoxy seal not only significantly decreased the time necessary to obtain a proper seal from

atmospheric conditions, but also had an overall greater aesthetic appeal. As shown in figure10

the grey Dow Corning seal looked unfinished and poorly applied whereas the clear epoxy sealant

leaves a clean finish.

5.5 Resistance Variable

An unknown variable was causing the tail end of the voltage vs. current graph to display

uncharacteristic behavior. Discussing these results with Professor Rastogi, he explained that the

main reason this characteristic in the plot was due in part to the large resistance being provided

by the carbon backing. Throughout this senior project a liquid carbon backing with an isopropyl

alcohol base would act as an electron donor for the potassium electrolyte solution. This carbon

backing had a resistance of approximately one to five kilo-ohms after application and drying

onto the conductive glass plate. This large resistance was believed to be the reason for the “tail”

of our I-V curves to have the result as shown in figure blah. Professor Rastogi suggested that we

heat our carbon coated conductive glass just the same as we would our titanium dioxide coated

conductive glass. The hope of this process was to dramatically reduce the overall resistance.

After approximately twenty minutes in the convection oven the resistance dropped to astounding

range of one hundred to three hundred ohms. This dramatic decrease in resistance led to a large

increase in current output.

5.6 Testing Voltage versus Current

Once the initial design was set, testing variations based on engineering judgment was crucial to

the overall direction of the analysis. By determining from the initial cells that a thin, even and

uniform dyed titanium dioxide would yield the best results.

The efficiency of a solar cell can be calculated by measuring the short circuit current, open

circuit voltage, and maximum power produced by the cell. Using a power supply and computer

data acquisition software, the current through the solar cell was measured for a range of supplied

voltages. The solar cell was tested both under darkness and illumination by tungsten-halogen

lamp. In this case, a blackberry stained solar cell on the larger conductive glass plates was used

because it yielded the largest measured current at over 3 milliamperes. The solar cell was tested

from the range of -1 volt to 1 volt at intervals of 0.1 volts, leading to 21 data points. Since

electron flow is forced through the solar cell, higher voltages cannot be used otherwise

components of the cell may be damaged. An efficient dye-sensitized solar cell naturally operates

between 0.4 – 0.7 volts; the testing voltage should be kept near this range.

Figure 21: Current (Amperes) versus Voltage (Volts) for Blackberry Solar Cell Without Illumination

Figure 22: Current (Amperes) versus Voltage (Volts) for Blackberry Solar Cell Under Tungsten-Halogen Lamp

Comparison of Figure 21 and Figure 22 shows the solar cell positively responds to light. In

Figure 21, the current goes through the origin at zero volts. In Figure 22, the current is much

higher at 4 mA at 0 volts. In general, the current is higher at each point in the solar cell under

tungsten-halogen lamp illumination. Each curve takes on a noticebly different shape. The cell

without light has a “S-shaped” curve because the current decreases with voltage, while the

illuminated cell has a concave shape because the current continues to increase over the span of

voltage. This shows the cell is producing power.

5.7 Calculating Efficiency

Figure 23: Current (Amperes) versus Voltage (Volts) for Blackberry Solar Cell Under Illumination

Focus on Maximum Power Produced As Shown By Rectangle

As shown in Figure 23, to determine the power produced by a solar cell, the rectangle with

maximum area under the curve in the fourth quadrant must be determined. It must be in the

second quadrant, where voltage is negative and current is positive, because this is where the

maximum power produced by a solar cell is located. The maximum power produced by the cell

can be calculated from:

where P is the maximum power, I is current, and V is voltage. The voltage and current are those

for the maximum area rectangle under the curve.

Pmax = 0.3 Volts * (1.96 * 10-3 Amperes)

Pmax = 5.9 * 10-4 Watts

A common measure of solar cell efficiency is the fill factor. The curve in Figure 23 takes on a

concave shape. The shape of the curve of an ideal solar cell will be a rectangle, with current

constant over the voltage range of the solar cell. The fill factor is a ratio which measures how

well this curve matches a rectangle in the ideal case. Fill factor is defined as:

where FF is the fill factor, Isc is the short circuit current, and VOC is the open circuit voltage. The

short circuit current is the y-intercept of the IV curve, while open circuit voltage is the x-

intercept.

FF = (5.9 * 10-4 Watts)/( 3.9*10-3 Amperes * .5 Volts)

FF = 0.30

The fill factor is low in this case. Solar cells with over 10% efficiency typically achieve fill

factors over 0.7. The low fill factor is mostly likely due to the crude blackberry dye used. Using

a ruthenium bipyridyl dye could result in a much higher fill factor.

Typically for solar cells, efficiency is:

where η is efficiency, PM is maximum power, E is the irradiance, and AC is the surface area of the

solar cell. E is typically taken to be 1000 W/m2 to simulate testing a solar cell on a bright day.

Using this measure, the efficiency of the solar cell is calculated.

η = (5.9 * 10-4 Watts)/(100 W/m * 0.0004 m2)

η = 0.0001475 = 0.1475%

This value of efficiency is low, considering typical silicon solar cells achieve efficiencies of

10%. However, dye-sensitized solar cells achieve lower efficiencies than silicon solar cells and

blackberry dye severely limits the efficiency. The use of a ruthenium bipyridyl dye would result

in efficiency over 5%.

In dye-sensitized solar cells, the efficiency is often measured using the following formula:

where iph is the integral photocurrent density, VOC is the open circuit voltage, FF is the fill factor,

and E is the intensity of light. The integral photocurrent density is the short circuit current

divided by the surface area of the cell.

iph = (3.9 * 10-3 Amperes)/(0.0004 m2)

iph = 9.75 Amp/m2

η = (9.75 Amp/m2)(0.5 Volts)(0.30)/(1000 W/m2)

η = 0.0014625 = 14.625%

This returns a similar efficiency to the other formula.

Conclusion

The titanium dioxide dye-sensitized solar cell is a very exciting product with real world

applications when further research increases the efficiency of these relatively inexpensive cells.

When creating these cells in large quantities it became clear just how many variables are

included in the sensitive chemical reactions which create energy. For example, the titanium

dioxide’s consistency was a major concern because of its delicate nature when sintering. The

acetic acid solutions needed to be tested to ensure they fell within the correct ph range. Our

sealant was scrutinized to ensure that no air could penetrate the cell, drying the electrolyte

solution which destroys the cell. The carbon backing needed to be prepared and heated to reduce

the high resistance. Should the long list of complications be addressed in one DSSC cell, all of

our energy needs could in effect be solved.

References

[1] Energy Information Administration, Electric Power Monthly U.S, Department of Energy (June 2005); available at hup://ww w.eia.doe.govJcncllfleleclrid ty/epmJchapJ.pdj

[2] Energy Information Administration, Electric Power Monthly U.S, Department of Energy (June 2005); available at hup://ww w.eia.doe.govJcncllfleleclrid ty/epmJchapJ.pdj

[3] Michael Grätzel (2006). The magic world of nanocrystals, from batteries to solar cells. Current Applied Physics: An official journal of the KPS, 6S1 e2-e7.

[4] Loucas Tsakalakos (2008). Nanostructures for photovoltaics. Materials Science and Engineering, R62 175-189.

[5] Michael Grätzel (2004). Conversion of sunlight to electric power by nanocrystalline dye-sensitized solar cells. Journal of Photochemistry and Photobiology, 164 3-14.

[6] Greg P. Smestad (June 1998). Demonstrating Electron Transfer and Nanotechnology: A Natural Dye-Sensitized Nanocrystalline Energy Converter. Journal of Chemical Education, 75 (6), 752-756.

[7] Chris X (?). Titanium Dioxide Dye-Sensitized Solar Cells

[8] Greg P. Smestad (June 1998). Demonstrating Electron Transfer and Nanotechnology: A Natural Dye-Sensitized Nanocrystalline Energy Converter. Journal of Chemical Education, 75 (6), 752-756.

[9] Michael Grätzel (2004). Conversion of sunlight to electric power by nanocrystalline dye-sensitized solar cells . Journal of Photochemistry and Photobiology, 164 3-14.

Appendix A: Requirements Matrix

The requirement matrix completed in the beginning weeks of the project was crucial to the most

important aspect of the design process, organization. Clearly defining each aspect of the entire

project for the instructors and advisors approval was an effective way of elaborating on the

processes taking place. The first semester saw many changes to the requirement matrix as the

project began to evolve and details changed.

Shall # Shalls Derived

Requirements

Derived - Derived

Requirements

Test Plan to meet

Req.

Shall #1

Build and study the components of a

Titanium Dye Sensitized Thin Film Solar Cell

Create nanofibers through various means to create solar cell, while

actively studying all its components

Efficiency must meet at least 2 - 5%

2. b. Use Voltage Testing

Program to Measure Current Output

Calculate Area Under Curve for Efficiency

Shall #2

Study, test and demonstrate the properties of the working solar cell

Using UV lamps and natural

sunlight, test working solar cell

for efficiency limits

Must try at least 3 various dyes

Experiment with different methods of

production

2.a. Digital Multimeter

Shall #3Create simulation

software for photovoltaic cell analysis

Design a simple software package

which will effectively test the

output voltage based on absorbed

sunlight. With emphasis on

sesitivity to UV light radiation.

Animation Software to Display Effect of

Wavelength of Absorption

3. b. Visual and Software

Check

Shall #4Research new materials

for design of new photovoltaic cell

With a more specific

understanding of a photovoltaic solar cells contruction

and internal reactions,

creatively research new, low cost

materials

3.c. Software Check

Shall #5 Simulation test of new materials

Using the created software package,

test the newly researched material's

properties for future build

considerations

2.b. Use Voltage Testing

Program to Measure Current Output

Calculate Area Under Curve for Efficiency

Shall #6 Test new design

Create the new material previously

researched and perform "real world" test for

efficiency in both laboratory and natural settings

(UV Lamp/Direct Sunlight)

Efficiency must meet at least 5-10%

2.b. Use Voltage Testing

Program to Measure Current Output

Calculate Area Under Curve for Efficiency

Appendix B: Block Diagram

The block diagram is a symbolic representation of the scope of this senior design project. By

simplifying the entire process into a few key steps, an objective observer can easily ascertain the

decision making and can give objective advice.

Appendix C: Budget

Materials Labor Acetic Acid - $10 • $100 per hour x 10 hours a week x 3 people Nitric Acid - $10 Total Labor = $3,900 Glycol - $10

Titanium Dioxide Powder – $30 Laboratory Space

Sealant - $75 • Economy Solid Tube Furnace - $1,560 Dyes - $120 • Digital Scale $125 Iodine - $40 • 400 ml beakers x 2 - $10

Total Materials = $221 • Mortar and Pestle - $26

Total Cost of Semester = $5,842.01

This is overall budget summary for the first semester. The senior project provided $100 dollars

per person, therefore the total budget was $300 for a group of three. In order to begin work on

the solar cell, initial supplies such as mortar & pestle and chemicals, were purchased at

Binghamton University Science Store. Sealant was purchased online, while the convection oven

and lamp were purchased locally. Conductive glass plates, titanium dioxide, laboratory space,

and synthetic dyes were provided by Professor Alok Rastogi.

Appendix D: Budget

Materials Labor Acetic Acid - $10 • $100 per hour x 10 hours a week x 3 people

Nitric Acid - $10 Total Labor = $3,900

Glycol - $10

Titanium Dioxide Powder – $30 Laboratory Space

Sealant - $75 • Economy Solid Tube Furnace - $1,560

Dyes - $120 • Digital Scale $125

Iodine - $40 • 400 ml beakers x 2 - $10

Total Materials = $221 • Mortar and Pestle - $26

Total Cost of Semester = $5,842.01

This is the entire budget breakdown from the first semester of the senior design project. The

materials were paid for by the allotted three hundred dollars from the class fund. All laboratory

space and supplies were provided by Professor Rastogi.

Appendix E: In lab pictures

1. Testing TiO2 layer sintering with the hotplate at 450 degree F; result was unsuccessful due to the heat could not concentration focus on the cell enough.

2. Testing TiO2 layer sanitizing by the Furnace at 450 degree F; result was successful.

3. Testing TiO2 layer sanitizing by the convection oven at 450 degree F; result was

successful.

4. Prototyping natural dyes from Black berry, Beet root and Pomegranate juice.

5. Prototyping standard dyes from Coumarin 153, Styrl 7 and Gallium (III) Phthalocyanine chloride.

6. Assembly with the liquid Graphite and insert the electrolyte solution and seal off to isolate the moisture from air.

7. Testing efficiency and compare the result by natural light or lamp and through computer program.