A dinner conversation the other day got me thinking: anthropogenic climate change—in large part due to increasing emissions of carbon dioxide—is happening rapidly (and increasingly inevitably), but how much do the various sources of carbon dioxide contribute to the problem? My dining companions and I anecdotally knew many facts about carbon sources: “Burning fossil fuels generates a lot of carbon dioxide”. “Transportation contributes, especially airplanes”. “Generating electricity must be a large source”. Thus our conversation went. But without a clear sense of the proportions of each of these carbon dioxide sources, we found it difficult to know where emissions reductions would come from.

That sounds like the kind of question that should be answered with a smart phone…or a blog post! Therefore, I present you with a pie chart, the perfect dessert item for this particular dinner conversation:

Pie chart showing the U.S. contributions to carbon dioxide emissions due to various sources. Data from the Inventory of U.S. Greenhouse Gas Emissions and Sinks.

A few things jump out at me: first of all, electricity and transportation make up more than 70% of all carbon dioxide emissions. Secondly, the 14% that industry contributes does not count electricity generation, so industry generates quite a bit of carbon dioxide independently of the energy required to manufacture products and/or to extract materials. Finally, we consumers are not without guilt, as we contribute to the carbon dioxide load just by living in our own homes.

Diving a little deeper, the chart above does not tell you anything about what kinds of transportation generate the most carbon dioxide. The Inventory of U.S. Greenhouse Gas Emissions and Sinks reveals that a massive 65% of all U.S. transportation emissions result from combustion in motor vehicles. Almost all the rest comes from burning diesel and jet fuel in heavy trucks and aircraft.

One final thought: reducing carbon dioxide emissions is not necessarily straight-forward. Suppose you decide to drive an electric car instead of a gas-powered vehicle: great! You just stopped burning fossil fuels on account of your transportation, but how is the electricity that recharges your car’s battery generated? If power generation does not also change to a cleaner source, then you have made less of a dent in the problem than you would like. At the dinner table, though, we did agree that riding a bike circumvents both problems and is a lot more fun than worrying about global climate change.

Okay, I admit, there are some ideas in energy that I run across that I get excited about out of proportion to their actual importance or current impact. Often, there is something diabolically clever or elegant about such an idea.

Case in point: using electric cars as storage for the electric grid. In her book, Before the Lights Go Out, Maggie Koerth-Baker discusses how the electric grid is ill-suited to handling large fluctuations in the supply or demand of electricity. Because neither solar nor wind power yield constant output, they will prove challenging to incorporate into the electric grid on a large scale. Further exacerbating the problem is the fact that not all conventional power plants can be quickly turned on or off in response to fluctuations in the supply or demand for electricity.

The solution to fluctuations in electricity is to implement large-scale storage of energy. There are a number of ideas for such storage, but Koerth-Baker points out that—and this is the diabolically clever idea!—a sufficiently large number of electric cars plugged into the grid could serve as a store of electricity. By adding or subtracting a small amount of electricity from the car battery of each electric car on the grid, fluctuations in the grid supply could be smoothed out.

What could make this feasible is that our cars sit unused, according to Koerth-Baker, 96% of the time. So as long as they are plugged in, the cars’ batteries act as a giant, collective battery that sops up any extra power on the grid or supplies some amount back to account for drops in renewable energy generation. The advantages of such a scheme are 1) no monolithic, large-capacity batteries have to be bought, at great expense, by an electricity utility and 2) the storage capacity of the network is more resilient.

While this is a scheme that has already been tested in Delaware, it will be a long time, if ever, before electric cars contribute in this fashion. Pre-requisites for the scheme include increased ownership of electric cars and the existence of millions of electric car charging stations that are connected to the electric grid. Nevertheless, it is a fascinating idea that I find myself (perhaps irrationally) excited about.

Neodymium. Praeseodymium. Terbium. Gallium. Europium. Lanthanum. Lithium. Samarium. Germanium. Dysprosium. What do these names mean to you? What relevance can they possibly have to your life? No, they’re not diseases you can contract, but rather examples of uncommon chemical elements that are all-important to renewable energy technologies in one way or another.

An underreported aspect of the drive towards renewable energies is the need for unusual and/or hard-to-find elements to be used in wind turbines (neodymium, dysprosium, praeseodymium, samarium), car batteries (lithium, lanthanum), solar panels (gallium, germanium), or even compact fluorescent light bulbs (terbium, europium). Many of these elements are categorized as “rare earth” elements, though they are not necessarily very rare. What makes them important is that since they constitute critical parts of renewable energy technologies, they will be needed in increasing quantities over the coming decades as we build ever greater numbers of solar panels, wind turbines, and electric car batteries.

The issues that these energy critical elements* have in common are related to extraction and geopolitics. Whether an element is considered to be “energy critical” may change over time depending on technology or circumstances. For example, until technical advances made aluminum production more efficient, aluminum ($1 per ounce in 1884) was as expensive as silver even though aluminum is the third most abundant element found in the earth’s crust. Further, consider what would happen to the electric car market should the country of Bolivia become less politically stable: with over half of the world’s lithium reserves, Bolivia is a place to pay attention to.

Extraction of energy critical elements can be difficult, both because of where they are located geographically and how they are produced geologically. Even when elements are relatively common, they may not be found in rich veins of high purity (e.g. germanium), thus making them hard and expensive to mine. Further complicating matters is that some elements are only the co-products of mining other, more common metals (e.g. gallium is extracted at the same time as aluminum and zinc), thus making for interesting economic trade-offs when trying to extract more of the less-common element.

Geopolitically, obtaining energy critical elements can be complicated when most of an element is found and/or mined only in one or two countries (e.g. neodymium is mostly mined in China, platinum in South Africa, and lithium in Bolivia). Such a monopoly on important elements enables market manipulation much like Saudi Arabia’s oil riches allow it the possibility of affecting oil prices around the world. In contrast, some countries will not even consider mining certain elements because the environmental impact is too high, and so mining of elements becomes concentrated in countries willing to tolerate these kinds of environmental costs.

Are there ways that we can mitigate the effect of a limited supply of energy critical elements? One of the simplest things is recycling energy critical elements from used equipment. In many cases, the purity of the metal in existing equipment is already very high making it advantageous to prolong the supply of the metal via recycling, However the cost of recycling can initially be high until enough material is recycled to create an economy of scale. Another solution is to support research into finding acceptable substitutes for elements which are being used up currently. Chemically similar substitute elements often can play the same role as the original element in making a functional device.

Ultimately, we need to recognize that our push to renewable energy will require obtaining the energy critical elements necessary to achieve our clean energy goals. Thinking ahead and planning for substitute elements will help us avoid falling back onto dirty energy technologies.

So tell me, smarty pants: if we’re not allowed to emit carbon dioxide by the metric ton anymore*, how am I going to be able to fly around the world without jet fuel? It isn’t as if I can stick a battery in my plane because there’s no battery in existence that contains enough energy for a long trip (except maybe here). Not to mention that if a battery fails in a Prius, I can probably pull over the car to the side of the road, whereas if the battery goes in my plane, then I am going to be in a world of hurt.

Biofuels are one possible replacement for liquid jet fuel. In this post, I will discuss why jet fuel is hard to replace and introduce biofuel as an alternative fuel. To begin, take a look at the following graph which shows the amount of energy found in different fuels:

Graph which shows the energy density of selected fuels. Fuels derived from crude oil have very good energy density, while batteries have poor energy density in comparison. Though biodiesel or ethanol-based fuels cannot solve all our liquid fuel needs, they could be a more environmentally friendly alternative for particular applications like jet fuel. Data from Wikipedia.

On the horizontal axis of the graph, you see how much energy there is per kilogram of material while on the vertical axis you see how much energy there is per liter of material. We care about energy density in terms of the mass because we want to avoid needing large quantities of fuel, and we care about the energy density in terms of the volume because we don’t want the fuel to take up lots of space (Yes, you should visit that link). On this graph, the best fuels in terms of energy density will show up higher up and more to the right.

From this graph, you can see that jet fuel is actually quite good because it packs a lot of energy into a relatively small space. Batteries barely even show up on the chart (bottom left corner) because they contain so little energy, thus demonstrating how ridiculous it would be to try to make a battery-powered airplane. So, any replacement for jet fuel has to have a lot of the same properties as jet fuel, but without the harmful environmental impact.

Let me get this out of the way right now: biofuels “cheat” to be considered an environmentally friendly alternative to jet fuel. Biofuels are chemically very similar to jet fuel and show up at a similar place on the graph I showed you (e,g, biodiesel, or the slightly less efficient ethanol), and therefore, they emit similar levels of greenhouse gases when they are burned for energy. They are better for the environment than jet fuels because of the way in which they are produced. Whereas fossil fuels contain carbon that has been sequestered in the ground for millions of years in the form of crude oil, biofuels are generated by growing plant material on the surface of the earth. The growing process itself takes carbon dioxide out of the atmosphere and creates a closed cycle for the carbon in the biofuel. To recap:

• growing a biofuel pulls carbon dioxide out of the atmosphere;
• burning the biofuel puts the carbon dioxide back into the atmosphere;
• growing the next generation of plants used for biofuel will pull the carbon dioxide back out.
• the net effect will be that a biofuel is carbon neutral, but that it will not be carbon-less.

Making biofuels requires growing plants. For example, ethanol can come from growing corn and fermenting the corn to turn it into fuel (kind of like how you brew beer…). Other sources of biofuel include fast growing switchgrasses or algae pools (see image below) which, according to one company, may even be able to produce ethanol without having to kill the algae. At the most basic level, however, you need to generate a large quantity of biomass that can then be converted into biofuels. This conversion process requires energy itself to happen, and the details of biofuel production can make a substantial difference in how cost-effective it turns out to be.

Photograph of algae pools being constructed in the desert. As with most other plants, algae uses the process of photosynthesis to turn carbon dioxide into hydrocarbons. Algae grows much faster than many plants, however, and can grow in pools of non-potable water. Using algae as the biomass for biofuels could help avoid tension between using agricultural land for food and non-food crops. Image courtesy of www.sapphireenergy.com.

In a future post, I will address more of the details of biofuel production, but here I want to finish by discussing some of the drawbacks of biofuel production. One problem comes from the fact that we have limited land on which to grow plant matter. A lot (or most?) of farmland should be dedicated to food production. If, instead, land is used to produce biofuels, we reduce the amount of food that can be produced for an increasing population, though some people claim (report forthcoming) that biofuels can be grown in places food crops cannot. Also, as the climate changes more over the next century, it will affect the amount of arable land**, probably leading to more conflict in how to produce environmentally friendly liquid fuels. Either way, if the demand for biofuels increases by ten or a hundred times the current levels, it is hard to believe that there will be enough land to generate the required biomass.

Ultimately, biofuels may only help us in limited ways. Rather than being looked at as a replacement for powering your car—where batteries or other technologies are better solutions—biofuels can be seen as an alternative to the otherwise-tricky-to-solve problem of lessening the environmental impact of air travel.

I received some comments regarding my earlier post on concentrated solar power that I think are worth sharing to the blog readership at large.

One technology that I did not know about is called a solar draft tower. The basic idea is that you build a really tall tower (more than 1000 feet, or about 300 meters, in most designs) and heats air inside a greenhouse area at the bottom of the tower. Then, that hot air rises through the tower, spinning turbines inside the tower.

The primary advantage of such a power plant is that it does not require very much water compared to most concentrated solar power designs. The main disadvantage is that the efficiency of such a structure is very low compared to other forms of solar power, turning only 0.5% of incoming solar radiation into usable electricity. For comparison, photovoltaic panels typically convert about 20% to 30% of the incoming solar energy to usable electricity.

To figure out the trade-offs between a solar draft tower and a CSP plant, I think one would need to do a life cycle analysis (LCA) of both approaches. I will write more about these in a future post, but in short, life cycle analyses basically account for all the production and construction “costs” in terms of energy, as well as all the outputs (good and bad) in terms of energy. Such an accounting of the two technologies would probably award big points to the updraft towers for their long lifetimes and low maintenance costs, but would give demerits for poor land usage and a higher environmental impact on-site. The outcome of such an analysis is unknown, however, without it having been done.

Water use was another topic that generated comments. Specifically, since water use is similarly high for CSP plants as for nuclear and coal plants, why would we choose to build any of these in the desert? I think this is a really good question that needs to be considered in building plants of any kind in places with low water availability. One alternative way of solving this problem is to use a heat transfer fluid other than water. For example, people have had some success using molten salt to serve the same role as water, but there are issues, such as a high freezing point for molten salts, which may need to be overcome for wider adoption.

Finally, in comparing CSP to photovoltaic panels, it was pointed out to me that CSP inherently contains more moving parts (to aim the light at the receiver) than a photovoltaic panel. Having more moving parts increases the cost and the complexity of CSP. On the other hand, many photovoltaic panels are also mounted on devices which track the sun and increase the overall collection ability of photovoltaics; moving photovoltaics could negate their cost advantage over CSP generation. For installation in populated areas, however, (non-moving) photovoltaics really seem to make the most sense because they can be modestly sized and reasonably priced.

Two presentations at the Energy Research Workshop were dedicated to talking about the prospects for nuclear energy in today’s world. The first made the case that the only way to meet our energy goals for the next 25 years is to build more nuclear energy plants while the second argued that small modular nuclear reactors are the best choice for new nuclear plants. In this post, I will focus on the latter topic, but first I want to think about nuclear energy more broadly.

Why do people get excited about nuclear energy? Because you want your energy source to have lots of energy for a very small quantity AND you would like your energy source not to take up too much space. Uranium-235, a commonly used nuclear fuel, meets both of these criteria. For instance, you would need 16,000 times more coal than Uranium-235 (by weight) to generate the same amount of electricity.

Nuclear energy is also appealing because climate change is a significant concern in today’s world and because many traditional sources of energy (e.g. coal and oil) spit out tons of pollutants which exacerbate climate change. In contrast, nuclear energy is a zero-carbon-emission source of electricity and does not directly contribute to climate change. (It is energy intensive to process into a usable form, however, and most of the uranium used in the United States is actually imported from other countries, so processing and transport indirectly contribute to climate change.)

So if the advantages of nuclear energy are so great, why are we not building nuclear plants as fast as possible? Because it is expensive. And secondarily, because there are safety concerns that surround nuclear plants. Of course, the two things are not independent of each other because necessary safety regulations do increase the cost of building and operating a nuclear power plant. The high capital cost of building nuclear plants means no new plants have been commissioned since 1975 and of those, only one has been completed in the last twenty years. Furthermore, if you look at the size of different power companies, investing up to $17 billion in a new, full-sized nuclear plant can mean betting the existence of the entire company on one project. That is a tall order and a big reason why nuclear is a hard sell without significant government subsidies.

Small modular nuclear reactors are a possible solution to the cost conundrum. Jose Reyes, the co-founder of a startup nuclear power company (sounds funny, right?) called NuScale Power, told us attendees of the Energy Research Workshop how, because of their modest size, small modular reactors could be assembled off-site at a factory and then shipped to the site of a nuclear power plant. Reyes claimed that making more reactors would reduce their individual cost because of economies of scale, but as this study shows, for the same amount of total generating power, small modular reactors have about the same price tag as large nuclear power plants. It is true, however, that a smaller reactor design also means that it is easier to build one or two at first—with the possibility of adding more reactor modules later—another cost-saving benefit for a company trying to build a plant from scratch.

Reyes also argued that situating the reactor cores underground would reduce some of the safety concerns associated with above-ground plants. First, situating the modules underground would help restrict access to the plants, a security advantage against 9/11-style threats. Secondly, reactors which are at risk of over-heating after an accident could have water added to them more easily by being below-ground. For instance, containment efforts at Fukushima were complicated by having to spray water onto the nuclear material containment vessels located on top of buildings, and small modular reactors might avoid this scenario in an emergency. Being below-ground is not always an advantage, however. Locating backup generators and other electronics below ground can increase the plant’s susceptibility to flooding and make it more difficult for crews to access the reactor in an emergency.

Cooling of nuclear material after it has undergone fission is always one of the biggest safety concerns to address with nuclear power plants. The argument in favor of small modular reactors is that passive cooling (simply by having either water or air around the reactor) is sufficient to safely contain nuclear material in an emergency. Since there is simply less material, then there is less heat generated and there is thus enough water around the containment vessel to bring the reactor down to a low enough temperature to be air-cooled. It is not clear, however, what would happen if multiple units in an area were damaged or if individual passive components (e.g. valves) lost functionality as the result of an accident.

Ultimately, the success of small modular nuclear reactors will depend on addressing safety concerns to at least the level of existing nuclear reactors. This process will also have to include satisfactory solutions to disposal of radioactive material, an issue that pertains to both large and small nuclear plants. If it costs too much to meet these standards, then it seems unlikely that small reactors will contribute to the energy landscape no matter their potential benefits.

Update (30 March 2012): Scientific American has posted a detailed article about small modular nuclear reactors. One point I did not make in my original post is that the military, particularly the Navy, has been using small nuclear reactors in submarines and on aircraft carriers for about 50 years with few to no significant accidents, so it can be done relatively safely with strict enough oversight.

One question kept reoccurring to me as I listened to the presentations at the Energy Research Workshop I attended yesterday: how much time do we have to switch over to new sources of energy?

In an ideal world, where we have plenty of time to change energy sources, we would choose the technological path that makes the most sense and has the fewest downsides, develop it, and then build it up to a level that could handle all of our power needs. So, for example, we might decide that wind power is best suited, on a large scale, to be our primary energy source. Even if it took 100 years to find all the materials needed to build enough wind turbines–some of the metals used to make wind turbines, like neodymium, are not found or mined in large quantities–we could afford to still emit pollutants from existing power plants and the cars we drive until the new generating capacity was complete.

But, instead, suppose we had only 20 years before climate change caused widespread and irreversible changes that affected most of earth’s population. Then, would we still choose wind power as the best solution to our problem if it took 100 years to complete construction of all those turbines? Of course not. Our answer would change depending on exactly how much time we had left and how quickly we thought different solutions would work. And, if we had very little time, we might even accept that some solution that, at first, seems like a bad idea, might turn out to be a better short-term solution because it saves us from driving off a cliff.

There are actually multiple goals that are driving us to change where we get our energy. For example, people might like to have a more sustainable source, one that will not run out in 100 years. In contrast, a country might also want to secure its “energy-independence” by being more self-sufficient in how it generates its energy. Or, we might need to change how it gets its energy to avoid disastrous changes to the climate that would make it more uncomfortable (or impossible) for us to live on the planet.

Each of these outcomes, however, comes with an associated deadline, and we may not know exactly what that deadline is. That is, if we continue with business-as-usual, will our sources of oil stay stable for 10 years or 50 years? Will we avoid the worst effects of climate change for 20 or 120 years? Trying to understand the amount of time associated with each of our goals (and there are certainly more goals than I have mentioned here) should make a big difference in how we approach energy research.

At the workshop yesterday, I noticed that the amount of time people thought we have left affected what they thought was the best solution to our energy problems. Sometimes, but less frequently, they got it backward too: what they thought was the best solution actually affected their estimate of how much time was left. In the next week or so, I plan to post my thoughts on particular topics from the workshop. But, underlying all of my posts is going to be an understanding that each energy technology has a set of assumptions built into it that affects whether it is an appropriate solution for whatever amount of time we actually have left.

Over time, I hope to develop a shared understanding with you, my reader, about which energy sources make the most sense to develop.

There is a method of generating electricity from solar energy that you may have never heard of. Rather than converting sunlight directly into electricity using a photovoltaic panel, concentrated solar power uses sunlight to heat water which, in turn, drives a turbine to generate electricity. How concentrated solar power works and whether it is a good method of generating electricity are questions I hope to answer for you.

Concentrated solar power (CSP) is actually an older technology than photovoltaic panels. In fact, the single largest existing solar generating plant, measured by generating capacity (470 MW), is based on CSP, and the first solar generating plant, built in Italy in 1968*, was based on CSP. These plants use a huge array of mirrors, covering square kilometers of area, to collect and focus sunlight onto water. At the input, liquid water enters and goes through the area where the sunlight is focused. As a result of the intense heat generated by the concentrated sunlight, the water turns into steam at the output. The steam continues through a complex of pipes to a turbine where it turns the blades of the turbine and generates electricity in much the same way as turbines at other types of power plants.

One of the advantages of a CSP plant is that its construction can be very simple**. No special materials are required for the collectors, which are just arrays of mirrors. Each array contains multiple mirrors and must be aimed at a receiver. The receiver itself is made of a material that absorbs and transfers heat efficiently to the water. For one CSP approach (the so-called “power tower”; see image below), these receivers may be up to 40 square feet in size. Furthermore, using steam to drive a turbine is a well-established technology, presenting no particular complications, and driving well over half of all the generating plants in the world (not just CSP).

Example of a power tower concentrating solar generator. Image credit: afloresm

So why, if CSP is the picture of simplicity, do we not hear more about it? In fact, Spain has perhaps the largest total installed solar power generating capacity through its numerous CSP installations. There is also a newer project to build one of the world’s largest CSP facilities near existing CSP installations in the Mojave desert. These do not come without an environmental impact, however, from covering so much of the desert with mirrors and needing to use large quantities of water. Water use associated with CSP generation is a particular concern in the desert where water conservation is already a big issue. Below, I have summarized Table 2 from this study of CSP water consumption (pdf), which compares how much water different power plant technologies consume.

Water is used for two purposes in CSP: the first is for the generation of steam while the second is the much more mundane need to wash the panels so they can still reflect sunlight at high levels. What is interesting here is that water-usage for CSP generation is largely competitive with other generation methods that are already used.

Ultimately, it is possible to reduce the water use of CSP generation if hybrid dry-wet cooling systems (pdf) are adopted, or if water is replaced by other transfer fluids like molten salt. Circulating molten salt through a CSP system would not only minimize the amount of water needed, but it could offer significant energy storage benefits (I will leave the details of energy storage for a future post).

So why would we choose CSP for power generation? I will answer this question in two ways. If we are comparing to all generation types, CSP offers an emissions-free alternative to coal- or natural gas-fired power plants–hence mitigating climate change–with an ability to keep costs reasonable. Should we propose it instead of photovoltaic solar power generation? Here I think it should not be a stark choice: CSP seems well-suited for large-scale installations because materials are not expensive and are easily replaced. Because PV panels often contain exotic materials which can degrade in performance over time, PV may be more difficult to turn into a large installation. On the other hand, it is a lot easier to install PV panels atop individual buildings, like your home, so that power generation is more distributed.

The final program (PDF) for the Energy Research Workshop has been posted. I am particularly excited that they have planned several “practical” sessions on how one takes ideas from concept to completion.

The Exploring Energies blog will officially launch on February 26th, as I attend the Energy Research Workshop organized by the American Physical Society.

Read the About Page to learn more about its purpose.