The Solar Cell (Photovoltaics)
The outer electrons of metal atoms can move through the body of the metal. They are said to be delocalized. The upper energy levels are called the conduction bands and the lower once the valence bands. The energy needed to excite electrons from the highest valence band into the lowest conduction band is called the band-gap energy. If a material has a large band-gap energy, the result is an insulator. If the gap is small, it is a semiconductor: some electrons have enough energy at room temp to jump the gap into the lowest conduction band. Metals become worse conductors with increasing temp because lattice vibrations inhibit electron movement; semiconductors become better conductors because more electrons gain enough energy to jump the gap.
The band-gap energy also corresponds to the frequency of radiation absorbed. The gap in diamond is large, so absorption is well into the UV part of the spectrum and diamond is therefore colorless. CdS has a band-gap energy which corresponds to a wavelength of 480 nm and it therefore absorbs blue-violet light and appears yellow. (I may not want to continue with this: materials seem to be able to absorb all colors corresponding to the band gap energy and more. It seems to me that once an electron gets up into the conduction band, it is delocalized and can go anywhere. For graphite, for instance, the BG energy is so low-corresponding to a wavelength of 700 nm, that all visible light is absorbed and it appears black.)
Electrons can not emit energy while orbiting or they would lose energy continuously and spiral into the nucleus.
The Semiconducting Diode
The 4 valence electrons of Si are used to bond it to its neighboring Si atoms. Small numbers of foreign atoms with five e in the outer shell (As or P) can fit into the Si lattice. But only 4 of their electrons are needed for bonding. Instead, it sits in the conduction band where it is free to move and conducts (drifts towards the + terminal) if an electric field is applied. This addition of foreign atoms is called doping. This is "n type" Si as it has an extra (-) charge. Typically, there is one P for every 104-107 Si. Using a trivalent atom like B, offers only 3 e for bonding. This leaves a hole in the valence band. It acts as a (+) charge carrier which moves through the valence band as valence e "drop into the hole" and another hole appears. This is called "p type" for (+).
Solar cells and semiconductor diodes are produced if p-type and n-type silicon are diffused together. At the junction the free electrons in the n-type are attracted to the "holes" in the p-type. They cross the junction and "fall into the holes". The n-type Si now becomes (+) in the vicinity of the junction. As the charge difference builds up there is a potential barrier which prevents further diffusion of charge (by repulsion). The area where there are no longer any free charge carriers is called the depletion layer.
Solar cells are typically doped Si devices which convert light energy directly into electricity. They are specially designed diodes where charge flow is affected by light. Photons of light hitting the solar cell are absorbed by electrons. This promotes the electrons into the conduction band and leaves a positive hole behind in the valence band. The electric field across the depletion layer causes the negative electrons to drift into the n-type material (which is positive now-see the previous paragraph) while the holes are attracted into the p-type. A potential difference exists between the two sides beyond the depletion layer. The cell can generate about 0.5-0.6 v, which will drive a current through an external circuit.
Semi-conductors like Si have a high refractive index. This means they are highly reflective. Less light is reflected (and more is transmitted) if the refractive indices of the two materials at the boundary are close in value. An anti-reflective coating is added to the top surface of a solar cell, this has a refractive index between that of air and Si, so that less light is reflected at the boundary. This makes the cell appear blue, as it is the blue end of the spectrum which is still reflected greatly (the RI of the coating for blue light is still quite different from the RI of air). The effect is to increase the amount of light transmitted into the Si by 30%.
Fuel Cells
The example we have is a PEMFC or proton exchange membrane fuel cell. Other fuel cells use other fuels and other ions to carry the charge. Many operate at high temperatures.
Making the Hydrogen
If H2 is to be used for storing energy which can be released in a controlled way by a fuel cell, it makes sense to produce the H2 by a method which uses up the least resources and produces the least pollutant waste. Hence the solar cell in our example.
Polymer electrolyte membrane (PEM) electrolyzers can break up pure water in almost exactly the reverse of the way in which a PEM fuel cell works. In a typical electrolyser cell the electrolyte is a thin membrane made for example of Nafion (a sulphonated polymer similar to Goretex, which is based on Telflon). It is only about 0.25 mm thick. The cathode has a porous C structure coated with very finely divided Pt; the anode has mixed ruthenium and iridium oxides as catalyst, again on a porous C base. (The anode support consists of Ti coated with Pt and the cathode support is C fiber. The C collectors carry current, and contain channels so that water can reach all the electrolyte and electrode surface.) Deionised water is used.
At the anode: 2 H2O ® O2 + 4 H+ + 4 e-
At the cathode: 2 H+ + 2 e- ® H2
Action of the Fuel Cell
In operation, the reaction occurring at the anode is: OT
H2 ® 2 H+ + 2 e-
The hydrogen ions pass through the membrane towards the cathode where they rejoin with the electrons that have passed through the external circuit:
O2 + 4 H+ + 4 e- ® 2 H2O
Reaction at the electrodes would be far too slow without a catalyst, which is usually Pt, which is a major factor in the cost. Research into reducing the size of the particles is ongoing, as that would decrease the mass of Pt used. The voltage obtained is 0.6-0.9 V direct current. Because they have very few moving parts compared to a conventional internal combustion engine, they can be very efficient in comparison-50-60% efficiency has been achieved in fuel cell engines for vehicles. The formation of water also generates heat which, if it can be recovered and used, raises the efficiency of the fuel cell considerably.
Other Fuel Cells and Their Uses
A prototype for a new fuel cell appeared in March, 2004. It is a Plexiglas cylinder with eight graphite rods around a hollow C and Pt tube. The rods serve as (-) electrodes, and the tube is the (+) electrode. Special bacteria stick to the graphite rods. Wastewater is pumped through the cell and the microbes break down the organic matter in a process that extracts e- from the waste and transfers them to the rods. They flow though a wire to the cathode. The protons left behind migrate through the wastewater to the cathode. Inside the hollow cathode tube, the protons combine with electrons and oxygen to generate pure water. So far the fuel cell can produce up to 150 milliwatts per square meter of electrode surface; they hope to get it up to 500-1000 mw. This would be enough power to pump the waste. It also needs to be made less expensive.
The Direct Methanol Fuel Cell:
It uses methanol directly, rather than breaking it down to release hydrogen. The reactions are:
Anode: CH3OH + H2O ® CO2 + 6 H+ + 6 e-
Cathode: 3/2 O2 + 6 H+ + 6 e- ® 3 H2O
Overall: CH3OH + 3/2 O2 ® CO2 + 3 H2O
The theoretical voltage is 1.21 V, but in practice is below 0.6 V. It has an efficiency of about 30%. Small ones are being developed for laptop computers and cell phones. You would only need to pour a little methanol into the chamber to create the electricity to run your device. (About 1 L would run a laptop for a week.) The CO2 released by a methanol fuel cell is the same CO2 that was taken up by the plant that produced the methanol in the first place. (This does not take into account the fossil fuels that go into the farming of methanol-producing plants.) It can come from sawdust or sewage, though, and that would be better.
There are other kinds of fuel cells that operate at high temperatures and are suitable for stationary applications-such as providing power to a building. In stationary applications, cogeneration is possible: using the "waste" heat that is given off when the water molecules are formed at the cathode. This gives an overall efficiency of around 80%. It is possible for stationary fuel cells to be larger and heavier than those in cars, so they are easier to make.
Any batteries used in a vehicle must be rechargeable, i.e. the chemical changes during discharge must be reversed. Batteries must also be light, occupy a relatively small volume, and be capable of being recharged rapidly. These three criteria, in spite of much research, are still far from being met. In addition, recharging batteries usually involves electricity generated in fossil fuel power stations, which is inefficient and also releases CO2.
If fossil fuel is used to generate H2 for a fuel cell, proportionally less CO2 will be released into the atmosphere, because of the higher efficiency of the fuel cells. Also, unlike a battery, a fuel cell needs no recharging.
To travel 500 km, a car requires 3 kg of hydrogen, which occupies 36,000 L at STP. It could be liquefied, but this is costly. Recently research has concentrated on absorbing the hydrogen into metal hydrides or carbon nanotubes. The former would require about 50 L of metal, and the later about 35 L, if research holds up. Rather than storing the hydrogen, some companies are working on fuel cells which produce their own hydrogen from methanol or gasoline, using an onboard "reformer." The reforming process does produce some pollution.
When looking at the cost of photovoltaics, remember that if the full environmental costs for the use of fossil and nuclear fuels (keeping the petroleum coming from the unstable Middle East, dealing with nuclear waste, global warming, etc), then solar power looks more economically sensible. Also remember that there was no infrastructure for the gasoline-powered car when it first appeared; the existing very well-developed infrastructure was for horses. It took about 20 years for one infrastructure to be replaced totally by another. The pace of the Internet and mobile telephone revolutions suggests tat there is now a shorter time scale for technological change.