Renewable energy: Power beneath our feet
8 October 2008
Magazine issue 2677.
by Julian Smith
AT FIRST glance, geothermal energy seems almost too good to be true. It's clean, inexhaustible, provides predictable 24-hour power and you can get it just about anywhere. A 2006 report by researchers at the Massachusetts Institute of Technology estimated that there is enough geothermal energy in the US alone to meet the country's energy needs 2000 times over. According to the Geothermal Energy Association (GEA), based in Washington DC, the best sites can generate electricity for as little as 5.5 cents per kilowatt-hour, compared with 8 or 9 cents per kilowatt-hour for natural gas plants.
There is a snag, however. Outside of geologically blessed places like Iceland, Japan and New Zealand, where volcanically active rocks are close to the surface, the Earth's heat is locked away under several kilometres of rock. Now, though, new developments are making these depths easier and more cost-effective to reach, and the world is beginning to realise the potential of geothermal energy.
The key to tapping this resource is a relatively recent technology called enhanced geothermal systems (EGS), which can create a geothermal hotspot pretty much anywhere. The process involves fracturing hot rocks, then injecting water, which heats up as it circulates through them. It is then pumped back to the surface and passed through a heat exchanger, which drives a turbine, generating electricity.
A number of EGS projects have recently come online. The world's first commercial plant in Landau, Germany, was commissioned in 2007 and already produces 22 gigawatt-hours of electricity per year. A 1.5-megawatt (MW) pilot plant in Soultz, France, began operating this June and a test plant at Groß Schönebeck, Germany, should be online by the end of next year. In southern Australia, a 1-MW demonstration plant should be producing electricity by January.
In the US, meanwhile, the Department of Energy has invested over $5 million to add an EGS system to a conventional geothermal well - where water is pumped through naturally hot rocks - east of Reno, Nevada, in the hope of increasing its productivity.
Sites like these are proving to be worthwhile investments, but that's mainly because, as existing oil, gas or conventional geothermal sites, their geology was already well understood. Some even had boreholes that could be adapted for EGS. Elsewhere, though, the costs of finding and tapping geothermal energy remain high.
Drilling, in particular, is costly - at Soultz it ate up 60 per cent of the investment. Clearly it is more cost-effective to drill where hot rocks are shallowest, but a lack of survey data means that these places are hard to find. "What they're mostly doing now is blind drilling where we see hot water coming out of the ground. It's the equivalent of the oil industry in the 1800s," says Karl Gawell of the GEA.
Geothermal is the equivalent of the oil industry in the 1800s
In fact, rocks at the necessary temperatures of between 150 and 250 °C are often 3 kilometres down or more. "The deeper you go, the more expensive it gets," says Jared Potter of California-based Potter Drilling. So, together with Jefferson Tester of MIT, Potter is developing a hydrothermal drill to try to change that equation.
The team has designed a system that replaces conventional drill bits with a high-pressure jet of steam at 800 °C. There is no solid cutting edge, so there is very little wear on the equipment. That means longer uninterrupted drilling times, fewer delays and significant savings. At the moment it costs up to $100 million to drill past 9 kilometres, says Potter. A 2006 study by Tester and colleagues estimated that reaching the same depth using the new drill could be up to an order of magnitude cheaper. "If these savings can be realised, EGS would become viable basically everywhere," Potter says. Even a savings factor of 2 or 3 could make it more widely workable. Potter and Tester hope to have a prototype ready for field testing next year.
Not everyone is convinced, however. "Our wells are more expensive than oil and gas wells because the product we're getting out isn't very valuable until we get a lot of it, and all at once," says Susan Petty of EGS development company AltaRock. The prices of the steel and cement used in well casings are soaring, she adds. "A new drill bit isn't necessarily going to fix this."
There are other ideas to make EGS projects more efficient. One is to replace the water sent through an EGS reservoir with "supercritical" carbon dioxide - CO2 at a temperature and pressure high enough to give it the properties of both gas and liquid. CO2 becomes sequestered in the rock in the process. The idea, developed by Donald Brown at Los Alamos National Laboratory, New Mexico, in the 1990s, would increase heat outputs because supercritical CO2 can move faster and more easily through the system than water. According to models this could increase heat extraction rates by 50 per cent and would sequester about 3.6 tonnes of CO2 for every megawatt-hour of electricity it produced. A state-of-the-art coal power plant produces around 0.8 tonnes of CO2 per megawatt-hour.
As well as financial considerations, there are practical obstacles to overcome before EGS is ready to go mainstream. For a start, injecting fluid into hot, dry rocks occasionally triggers earthquakes. An EGS project in Basel, Switzerland, was suspended in 2006 after it triggered a 3.4-magnitude quake. Researchers are trying to understand what exactly causes these quakes and how to prevent them.
In Iceland, by contrast, the viability of geothermal energy isn't a problem - over half the country's energy comes from geothermal sources. Even so, one project is hoping to prove that there's no such thing as too much green energy. The Iceland Deep Drilling Project (IDDP), which began in 2000, aims to take geothermal beyond steam by drilling into a reservoir of water which has been heated to 450 °C by a magma chamber below and is in a supercritical state. It's tricky stuff to handle - generally geothermal engineers try to avoid these reservoirs as they can cause dangerous and expensive blow-outs. "Supercritical fluids are not very predictable - there is the risk of explosion," says Ragnar Asmundsson, coordinator of the HITI project, a spin-off of the IDDP that is currently drilling towards one of these unusual reservoirs. The risk may well be worth it, though. If they can tame the fluid, the wells could each produce an order of magnitude more power than a conventional geothermal borehole.
Similarly explosive geothermal conditions can be found in underground reservoirs where gas-saturated liquids are trapped in deep sedimentary formations. These "geopressured" reservoirs could yield not only heat from the fluid itself, but also chemical energy from the dissolved natural gas, and hydraulic energy from the extreme pressure. Germany and Australia are currently looking into ways to exploit these resources.
Interest in geothermal is at an all-time high, but while there is no shortage of heat, the same can't be said for funds. A healthy injection of investment - and government subsidies - is desperately needed if geothermal power is to achieve its potential. In the US, the Department of Energy has proposed the creation of a national geothermal database to help interested parties strike heat. This August, Google.org, the charitable arm of the internet giant, announced a $10 million funding package for projects to map geothermal resources or make drilling more cost-effective. There's still a way to go, but according to Dan Reicher, director of climate and energy initiatives at Google.org, geothermal - and EGS in particular - is going to be big. "[It] could be the 'killer app' of the energy world," he says.
Enhanced geothermal could be the 'killer app' of the energy world
Sunny side up
There's much more to solar power than photovoltaic cells
Photovoltaic cells are currently the fastest growing energy technology, with production increasing by around 48 per cent each year. By 2015 the price of electricity from PV cells is expected to match that of conventional energy generators. (For a detailed account of the state of the art in photovoltaic cells see New Scientist, 8 December 2007, p 32.)
But photovoltaic cells aren't the only way to capture the power of the sun. Large-scale concentrating solar power (CSP) systems are all the rage in the energy-hungry US. Last June, the 64-megawatt (MW) Nevada Solar One CSP plant switched on near Boulder City. Since then, over 1.6 gigawatts of new CSP capacity have been announced in neighbouring California.
In the past year or so, the US Bureau of Land Management has received more than 30 planning requests to develop large-scale CSP plants across the US. The situation is similar in Europe, where around a dozen plants are under construction with at least 24 more proposed in Spain alone.
Sound economics lie behind this enthusiasm. At the moment, electricity generated by large-scale solar concentrator systems costs around 12 US cents per kilowatt-hour. Though this is around four times the price of electricity from a coal-fired power station, it's half the price of electricity produced by photovoltaic cells. What's more, this technology offers an advantage that could prove decisive in the longer term: the ability to store energy for hours or days at a stretch.
Rather than converting sunlight directly into electricity, a CSP system uses arrays of mirrors to focus sunlight onto tubes filled with water or oil. The fluid is heated under pressure to around 400 °C and is then circulated to a steam turbine to generate electricity. By replacing the water or oil with molten salts, typically a mix of sodium nitrate (NaNO3) and potassium nitrate (KNO3), and storing this hot mixture in insulated tanks, it is possible to use energy collected during daylight hours to generate electricity at times of peak demand - day or night.
"With this system you can make electricity when you want," says Massimo Falchetta, an engineer at the Italian National Agency for New Technologies, Energy and the Environment, in Rome. That means the energy can be sold for a higher price than stuff from wind generators or PV cells.
Solar concentrators are nothing new. The Solar Energy Generating Systems (SEGS) plant has been operating in California's Mojave desert since 1985. Made up of nine energy farms capable of generating a total of 354 MW, SEGS is the largest solar concentrator system in the world.
During the 1980s and 1990s, US engineers tested various designs, including power towers - mirrors arranged around a vertical pipe system - and systems with molten salts. However, the US Department of Energy halted research in 2000 after the US National Research Council suggested that any further gains in performance would be insignificant.
Development continued in Europe, and later this year the first commercial molten salt-based solar collector system is due to be switched on at Guadix in Andalusia, Spain. Andasol-1 has over 500,000 square metres of parabolic mirrors and will generate 50 MW of power. With large storage tanks for the salt solution, it will be able to continue generating electricity for more than 7 hours after sunset.
In April, the Electric Power Research Institute in California released a report suggesting that adding up to 9 hours of energy storage with molten salts to a solar concentrator plant can reduce the cost of its electricity by up to 13 per cent.
This cost could fall further if new experimental fluids containing nanoparticles outperform salts, says Mark Mehos, who manages the solar thermal power programme at the National Renewable Energy Lab in Golden, Colorado. "It's early days but this has the potential to be revolutionary," he says.
With so much plant waste around, it should be easy to bring biomass into the mainstream. So what's the problem?
Biomass is plentiful enough. You can burn wood, manure and more or less any other kind of plant material. Indeed, it provides around 11 per cent of the world's energy, mainly as heat, although in most industrialised nations it yields just 1 or 2 per cent of their electricity at best. And despite a growing enthusiasm for all things green, biopower faces some distinct challenges if it is to move from the margins to mainstream production.
The problem is, biomass doesn't actually yield much energy. Felled trees and energy crops are only a quarter as energy dense as bituminous coal - biomass generally yields around 7 gigajoules per tonne - so right now biopower is struggling to break even economically. In Sweden, which generates around 4 per cent of its electricity from biomass, the break-even figure for willow plantations is 12 to 16 tonnes of dry wood biomass per hectare each year. A recent study at Lund University suggests that only along its wet western coast can Sweden exceed this, with annual yields of up to 17 tonnes per hectare.
Yet biomass does have some advantages. For example, Canada's forestry industry faces a serious threat from the mountain pine beetle, a pest that has already devastated more than 130,000 square kilometres of Canadian forests (New Scientist, 18 December 2004, p 16). Beetle-infested timber is useless as lumber, but it could be good news for biopower, argues Amit Kumar at the University of Alberta in Edmonton. The infested wood could support three 300-megawatt (MW) power stations, providing around 1 per cent of Canada's electricity needs.
It's telling that Canada, with over 400 million wooded hectares, has no healthy forest to devote to biopower. The lumber and paper industries monopolise it and, as Kumar suggests, biopower can succeed only by feeding on their scraps. This problem is even more apparent elsewhere. In the US, biopower has become the single largest provider of renewable electricity, thanks primarily to waste biomass from the paper industry. That resource is now fully exploited, though, and the US Department of Energy estimates that future biopower growth will slow, providing around 1.7 per cent of the country's electricity by 2030.
So where can we find biomass to feed the power stations? Fortunately, untapped resources exist in many parts of the world, including waste from sugar-cane processing in Brazil and waste palm oil and rice hulls in Asia. In the US, livestock produce over 1 billion tonnes of manure annually, most of which is left to rot. Why not collect the stuff, convert it to methane in anaerobic digesters and use it in power stations? Michael Webber at the University of Texas in Austin calculates that US manure could generate 68 million megawatt-hours each year - almost 2 per cent of the annual US electricity demand (Environmental Research Letters, vol 3, p 034002). The infrastructure to collect and transport manure would incur its own costs, so local-scale operations might be key to maximising profits.
With that in mind, plans by Finnish company Wärtsilä to build small combined heat and power plants attached to two breweries in the north of England could be a sign of things to come. Using spent grain from the brewing process to generate power, their output will be modest - around 3.1 MW of electricity each. That's plenty to power the breweries' operations, though, and any excess will be sold to the grid.
When electricity has to travel thousands of miles you need a different kind of grid
Sometimes newest isn't best. A technology dismissed as obsolete a century ago could turn out to be the key to building a power grid fit for delivering electricity generated from renewable sources.
Nearly all of the world's power lines carry alternating current (AC). The reasons for this go back to an epic argument in the late 19th century between two of the biggest names in the history of electricity: Thomas Edison, the inventor of the light bulb, and the engineer Nikola Tesla. Edison argued that direct current (DC) was the right way to transmit power over long distances, because with AC this can only be done if the voltage is stepped up to lethal levels. He even built the first electric chair to demonstrate his point. Edison lost the "war of currents" because Tesla's AC transmission system proved more practical, and so it remained through the 20th century.
But Edison may yet be vindicated. Unlike conventional power plants, which can be built close to where their electricity is needed, renewable energy sources are not always near population centres, so this power must be transmitted over long distances. Over a distance of 1000 kilometres, AC transmission lines become increasingly inefficient, losing over 10 per cent of the energy pumped into them, whereas a high-voltage DC line would lose just 3 per cent. When you factor in conversion from DC to the AC supply that consumers need, the additional losses are 0.6 per cent at most. In all, at distances of about 800 kilometres and above, DC transmission becomes cheaper than AC.
This has led supporters of renewables to call for a new generation of high-voltage DC "supergrids" linking regions rich in wind, solar or other renewable energy sources with populated areas thousands of miles away.
In May, the environmentalist Robert Kennedy Jr called on the next US president to build a high-voltage DC grid to transport electricity from the "wind corridor" that stretches from the Texas panhandle to North Dakota, and solar power plants in the Southwest, to all major US cities.
In Europe, Gregor Czisch, an energy systems expert at the University of Kassel in Germany, has calculated the costs and benefits of a supergrid stretching from western Siberia to Senegal and providing 1.1 billion Europeans with 4 million gigawatt-hours of electricity a year from renewables. The grid would link onshore and offshore wind farms, hydropower resources and solar concentrator power plants with major European cities. The sheer extent of a grid like this would do a lot to smooth out the energy supply, Czisch says.
Not only would the electricity this provides be clean, it would also be reasonably cheap. At 2007 prices, a European supergrid would deliver wholesale electricity at a cost of ¤0.047 per kilowatt-hour, says Czisch, compared with ¤0.06 to ¤0.07 per kilowatt-hour for electricity from gas-fired power plants.
For the latest from New Scientiist visit www.newscientist.com