Solar Energy Research
As the world's energy consumption rapidly grows, we are exhausting finite stores of fossil fuels, but more importantly, we are emitting massive amounts of carbon dioxide and other greenhouse gases that contribute to global warming. It is becoming more and more urgent that we develop viable alternative sources of energy. I am interested in particular in solar energy: converting sunlight into electricity and fuels such as hydrogen. Why solar? Mostly because it is so abundant: more energy from sunlight strikes the earth in a single hour than the entire world uses in a year!
There are three ways to convert sunlight into energy, shown in the diagram below, each with advantages and disadvantages. Nature’s method of choice is photosynthesis, shown on the left. Photosynthesis occurs in the leaves of plants, where sunlight and water are used to produces oxygen and sugars (what the plant uses to grow and sustain itself). Photosynthesis can only produce chemical fuels, and of course does not produce electricity. Photosynthesis is by far the largest consumer of solar energy, but the process isn’t very efficient in converting sunlight to usable energy (estimated at ~0.5%). As humans, we use the energy produced by photosynthesis both from burning wood from trees we’ve cut down ourselves, but also from burning fossil fuels, such as oil and coal, which were formed from plant matter that grew millions of years ago.
By far, the most efficient way to convert sunlight to energy is with photovoltaic solar cells, shown on the right in Figure 1. These solar cells are made from various different kinds of semiconductors, such as silicon, gallium arsinide, and indium phosphide among others. Photovoltaic devices are almost one hundred times more efficient than photosynthesis at producing usable energy from sunlight, but their price tag is quite a bit higher... hundreds of times higher. These are the solar cells you might find on somebody’s roof, or a solar-powered calculator, on satellites, and at some solar power plants. These have been engineered so well that little improvement in efficiency is even possible, but much work remains to be done to reduce their cost. Also, these units only produce electricity, and cannot produce fuels. They have no inherent ability to store the energy they produce, which is a problem at night! Unfortunately, electricity is a very difficult form of energy to store without loosing a lot of the energy in the storage process. Also, keep in mind that electricity is not always the most useful source of energy – the large majority (~85%) of energy used in the world today is from fuels, such as coal, oil, and natural gas, and not electricity.
The third way to convert sunlight to usable energy is in between the first two in all respects: it makes both fuels and electricity; it is more efficient than photosynthesis but less so than photovoltaics; and its price is also somewhere in between. These are known and semiconductor-liquid junction solar cells. They are generally much easier and cheaper to fabricate than photovoltaics. Because they can directly make fuels, such as H2, and much more efficiently that photosynthesis can, they are very promising as emission-free alternative energy sources. Maximizing their efficiency while keeping their cost low is the name of the game here! Semiconductor-liquid junctions are what I study here at Caltech.
The research I did for my Ph.D was on a new type of solar cell, which is a unique kind of semiconductor-liquid junction, and is shown schematically below. It used a thin film of nanometer-sized particles of the semiconductor titanium dioxide (TiO2). Unlike many semiconductors, TiO2 is very inexpensive and abundant on earth. For instance, it is commonly used to make white paint white! You can also find it commonly in sunscreen, and toothpaste, among other things. Unlike silicon, which is black, TiO2 does not absorb very much sunlight directly (it is white). Therefore a dye is bonded to the surface of the TiO2, and it is this dye that absorbs the sunlight instead. The dye consequently transfers the energy it has absorbed from the sunlight to the TiO2, and electricity is then produced. The most common dye using the metal ruthenium, or Ru. The dyed TiO2 film is immersed in a liquid solution that contains the element iodine in two different form, known as iodide and triiodide, or I- and I3- respectively.
Figure 1: Three methods for directly converting sunlight into usable energy.
This design is very promising in that it is fairly efficient and is cheaper to produce than silicon-based solar cells. In fact, I can make a working solar cell in just a few minutes in my lab! However, more than 15 years after being developed, the individual steps and reactions occurring in the cell are still not fully understood. We predict that there is still room for improvement in the design, leading to more efficient conversion to electricity. It is of course difficult to optimize the design of something that is not very well understood. My research is focused on better understanding and characterizing how the cell works and how to directly alter and control the properties of the solar cell. While my research is mostly regarding questions of fundamental science (as opposed to engineering and optimizing efficiency), the context for the questions we seek to answer is to improve the efficiency of the solar while keeping the cost of production as low as possible. The system is complex and the individual reactions we seek to understand are highly interdependent.
Figure 2: A schematic diagram of the components of the TiO2-based dye-sensitized solar cell, with a Ru-based dye and using I3-/I- as a redox couple.
page last updated 10/09