Introduction: How to Build & Use a Dye-Sensitized Solar Cell (DSSC) + a Discussion on Energy & Efficiency

Harnessing renewable energy sources is crucial for supporting the energy demands of modern society. The IEA calculated that global energy usage increased by 10% from 1990 to 2008, and the number is expected to rise in the coming decades. At the same time, our currently predominant sources of energy pose serious threats to the environment and human health. The burning of fossil fuels—natural gas, coal, and oil—releases harmful greenhouse gases into the atmosphere, contributes to ocean acidification, and creates pollution with both economic and social costs.

Solar power is a particularly promising solution to the world's energy needs. The Earth annually absorbs nearly 4 million exajoules of solar energy, and it would require less than an hour of this total energy to power mankind for an entire year.
There are many technologies to capture and convert the sun's energy. My research team has been experimenting with DSSCs: dye-sensitized solar cells. They differ from traditional photovoltaic (PV) cells that currently dominate the solar cell market, and DSSCs have their own advantages and disadvantages. The tradeoffs will be discussed later in this guide. (View the ground-breaking Smestad and Grätzel paper "Demonstrating Electron Transfer & Nanotechnology: A Natural Dye-Sensitized Nano-Crystalline Energy Converter" to learn more about the technology).

I will be explaining how dye-sensitized solar cells are assembled, and how they can power an electronic gadget---in this case, a calculator.

Step 1: Gather Your Materials

Here are the necessary materials for this project:
- Fluoride-doped tin dioxide conductive glass (≥2 plates)
A source of organic dye (ex: raspberries)
- Titanium Dioxide Paste
- Acetic Acid
- Water
- Glass Stirring Rod
- Scotch Tape
- Graphite Pencil
- Liquid Electrolyte Solution
- Binder Clips
- Wires
- A Solar-powered Calculator (You will be removing the stock PV solar cell and replace it with your assembled DSSC)
- Soldering Equipment
- A Multimeter or Vernier Lab Quest (with Voltage & Current Probe(s))
- Tungsten Halogen Lamp
- Testing Apparatus

Step 2: Prepare the DSSC Electrodes

For every solar cell you assemble, you will need an anode and a cathode. The anode will contain the dye and titanium dioxide molecules. Photons will excite the dye molecules' electrons, and the electrons will jump from the dye molecule to the titanium dioxide to the glass anode through diffusion. The electrons will subsequently travel through a circuit that connects the two electrodes and return to the cell through the glass cathode. To complete the circuit, an interstitial electrolyte solution facilitates the flow of electrons back to the anode's organic dye molecules.

So, to assemble your anode, you must:
1) Prepare a titanium dioxide solution

Slowly add 20mL of an acetic acid solution (0.1675 mL CH3COOH per 99.8225 mL water) to 12g of titanium dioxide powder. Slowly adding the acid, in addition to vigorously mixing the solution, will ensure a uniform paste. It is recommended that you use a mortar and pestle for this step. (Figures 4 & 5)

2) Anneal titanium dioxide to your first glass plate

Cover each of the FTO glass plate's four edges with a 2mm-thick piece of scotch tape (on the conductive side, as determined by a multimeter). This will create an ultra-thin "bowl" that will be filled with the titanium dioxide solution. Apply three drops of the TiO2 solution to your electrode, and spread evenly and gently with a glass stirring rod. (Figures 3 & 6)

Anneal the glass plate with an oven (or other heat source) at 450C for 30 minutes. (Figure 7)


3) Soak the annealed electrode in your dye solution

Create a solution that contains dye molecules. This is done most easily with the juice from frozen raspberries (which can easily be purchased in the frozen foods section of a supermarket). The organic dye solution can be purified through filtration methods (Note, however, my research team used very basic methods. We pulverized frozen raspberries into a juice with a mortar and pestle. Afterwards, we squeezed this solution through a cheesecloth, and this process removed the largest pieces of  pulp.)  Regardless, soak the annealed electrode in your dye solution for ten minutes. (You should notice by then end of this period that the color of your glass plate has changed as dye molecules covalently bonded to the TiO2. This is the process known as sensitization. You should also be aware that other dye solutions might require soaking for different lengths of time) Gently rinse off the glass plate with water once and ethanol twice so that remaining pulp and sugars are removed. (Figures 8, 9, and 10)

As for your cathode:
1) Cover one side of the glass surface with a carbon film (using the graphite pencil).

This process is rather self-explanatory. Just make sure that you are applying a complete coating of graphite to the plate. This may take time depending on your pencil. (Figure 11)

2) Optional: Anneal the cathode in an oven at 450C for a few minutes.

3) Gently rinse the cathode with ethanol.

Step 3: Assemble Your Solar Cell

Awesome! With an anode and a cathode, you've almost finished constructing the DSSC.

To assemble this DSSC, place the carbon-coated side of your cathode on top of the dye-coated side of your anode. Do not allow your glass plates to overlap completely, offset the electrodes so that they fully line up horizontally but extend beyond each other's vertical ends. The offset will allow alligator wires to clip into each of the electrodes, thus allowing you to assemble a circuit. Secure your offset glass plates with binder clips on the two horizontal sides that line up completely. (Figure 13)

To insert the electrolyte solution between the two electrodes, apply the electrolyte liquid along an offset edge in between the two glass plates. Then, open and close the binder clips in an alternating fashion (one binder clip at a time, not both). This will cause the electrolyte solution to be sucked up into the cell and be distributed evenly among the graphite coating and TiO2/anthocyanin complex. (Figure 12)

Step 4: Test Your Solar Cell

Connect your assembled DSSC to a multimeter via alligator clips and wires. After the assembled circuit is complete, place your cell under an indoor light source---or along a windowsill. (Warning: Placing your solar cell directly under the sun without the proper protection will cause severe degradation. Do initial testing under light sources that do not emit UV radiation). Record the maximum voltage and current produced by your solar cell. Multiply these two figures to obtain the power output.

Power is simply the voltage potential multiplied by the current. So, for example, if my DSSC produced 0.400 Volts and 250 microamps (0.000250A), the power output would be 0.1 milliwatt (0.0001 watt).

Visit AllAboutCircuit's guide to learn more about circuits, voltage, current, and power.

Step 5: Modify Your Calculator

The utility of an assembled solar cell is found through powering electronic devices, with energy that has either been produced in real-time or previously stored. I will demonstrate how I powered a calculator with the cells that I produced. However, with some customization, you could power a cornucopia of electronic devices. Charging (rechargeable) batteries could be extremely useful. If you wish to follow along my guide, try powering a calculator for now.

It is preferable to find a calculator that is already solar-powered, for it will have the existing circuitry that will allow it to be easily powered with your DSSC. Remove the external casing with the necessary tools so that you may have access to the internal circuitry. Solder off the wires that connect the calculator's built in solar cell to the circuit board and solder in two new wires that will connect to your DSSC. 

(If needed, view this useful tutorial for information on soldering)

Before connecting the two new wires to the DSSC, connect the calculator to a variable power supply. Determine the minimum voltage and current needed to power the calculator, and ensure that these demands can be satisfied by your solar cells. If not, consider producing more solar cells. Connecting multiple DSSCs in series would boost the voltage, while connect them in parallel will boost the current. If these methods still prove insufficient, you could charge batteries or capacitors with your DSSC(s), and then power your calculator with the stored energy. 

View the circuit I created to power a calculator here: http://youtu.be/g1KwfynkDIU

Step 6: Discussion on Energy: Efficiency

As an emerging technology, dye-sensitized solar cells perform poorly when compared to either traditional solar cells or fossil fuel energy sources. Although the sun is a plentiful source of energy, many factors limit DSSCs' efficiency of converting sunlight into electrical energy. (View the article "Advancing beyond current generation dye-sensitized solar cells" for more information about DSSC's limitations and potential avenues for improvement).

While fossil fuels have many drawbacks, their advantages include energy density and low cost (without considering externalities). To best demonstrate the advantages and disadvantages of alternative energies, it is best to examine this issue through the lens of powering the calculator from earlier in this guide.

To power this calculator with our DSSCs, we connected three cells (each approximately 3 cm^2 in surface area) in series to the calculator. Our circuit produced 1.04V and 630 microamps, i.e. 0.6552 milliwatts of power. The cells had been placed on a windowsill, receiving a moderate amount of sunlight during the early afternoon of a New England winter day. According to the aforementioned Grätzel paper, one could expect 600-800 W/m^2 (0.54-0.72 W/ 9 cm^2) from sunlight to be reaching the solar cells.
To find the amount of energy produced by the calculator's original solar cell, I also connected the PV cell to a multimeter while it rested on the same windowsill. However, the PV cell and DSSCs were measured on different days with different amounts of available sunlight, so this is not a perfect comparison. Nevertheless, it is reasonable to expect only small variations between two winter afternoons in New England that were less than one week apart. The PV cell produced 9.2 milliwatts of power (2.8 milliamps and 3.27 volts). This is 14X the performance of the 3 DSSCs, and 21X the performance of the DSS cells per unit of area. (15.20661157 W/m^2 for the PV vs 0.728 W/m^2 for the DSSCs).

To find the amount of a conventional energy source needed to power this calculator, I know that a kilogram of coal (typically the most energy dense fossil fuel) will approximately produce 7.4 megajoules of electricity. Therefore, one would need 2.92*10^(-7) kilograms (i.e. 0.292 milligrams) of coal to power this calculator for every hour used.

These numbers show the inherent difficulties with converting solar energy into electricity. I set out to calculate an efficiency measurement for the DSSCs that were produced, in addition to the photovoltaic (PV) solar cell that originally powered the calculator. Both cells were tested inside a cardboard box apparatus where a single light source (45W 120V tungsten halogen lamp) shone on the solar cells at an equal distance from above. There were no other light sources that entered this apparatus, and the inside was spray painted black to minimize light reflection from the cardboard.

One of the DSSCs produced 5.5 microwatts of power, or 0.01897 Watts/m^2 when tested inside the box. This power output per unit area was significantly lower than the original calculator cell’s 1.49 Watts/m^2 (2.99V * 301 microamps / 6.05 cm^2). Considering that the DSSCs had 1.27% the efficiency of a cheap, mass-market calculator’s solar cell, the capabilities of basic anthocyanin-based DSSC technology left much to be desired. The original Grätzel paper mentioned that one could expect efficiencies between 0.5% and 1% for DSSCs constructed in their experimental procedures. This creates confusion as to why the PV outperformed the DSSC 78X per unit of area. Based on current technologies, it would be impossible for an inexpensive calculator’s PV cell to create electricity with an efficiency between 39% and 78%. This discrepancy could be the result of a tungsten halogen lamp emitting rays of light different than those of the sun. The DSSCs did in fact perform more comparably when tested along the windowsill. It is also possible that our research team did not create cells of the same quality and efficiency as those who wrote the original Grätzel paper. 

There are some other important considerations to these comparisons of energy sources. First of all, dye-sensitized solar cells have issues of stability that would most likely prevent them from operating for twenty-plus years (the typical lifespan of a PV cell). My team has noticed significant performance degradation to our cell mere weeks after assembly; however, we have not taken many of the possible precautions in extending the cell's lifetime. (Read about one research team's efforts with enhancing the outdoor stability of their DSSCs) Nevertheless, this limitation highlights one problem with current renewable energy production. There are some technologies that have the opportunity to reasonably compete with coal and other carbon-based energy sources, but many technologies have serious financial and technological roadblocks that prevent mainstream adoption. Based on my research team’s experiments, photovoltaic solar cells currently hold much more potential than dye-sensitized ones at powering the world’s future energy needs.