The Economist, December 2, 2006

CAN COAL BE CLEAN?

[Rachel's introduction: As the end of the petroleum era peeks over the horizon, the chemical industry and the energy industry are planning to shift to "clean" coal. Unfortunately, there is no such thing as "clean" coal.]

Two new coal-fired power plants will soon appear on the banks of the Ohio River if American Electric Power (AEP), a utility, gets its way. There is nothing unusual about that, of course: a rival firm, TXU, unveiled plans to build 11 new coal-fired plants in Texas earlier this year. All told, there are some 150 plants on drawing boards around America, which derives 56% of its power from coal. But AEP's two plants are different in one critical respect: their design would make it relatively easy to filter the carbon dioxide out of their emissions, should the company ever need to do so. America's first "capture-ready" power plants, to use the industry parlance, are nearing construction.

Coal has several advantages as a fuel. It is abundant. It is widely distributed: countries that are short of other fossil fuels, such as Germany and South Africa, have mountains of it. As a result, it is cheap. Even though the price has risen in the past few years, it is still less expensive to run a power plant on coal than on almost anything else.

But coal is also dirty. It releases lots of soot and various noxious chemicals as it burns, and so has fallen out of favour in many Western countries. Worse, coal-fired plants produce roughly twice as much carbon dioxide per unit of electricity generated than those that run on natural gas. Power generation contributes more to global warming than any other industry, and coal is the dirtiest part of it. Coal- fired plants are responsible for perhaps 8 billion out of the 28 billion tonnes of man-made carbon dioxide released every year, and are thus a prime target for emissions cuts. If environmentalists had their way, there would be no coal-fired plants at all: protesters recently called for the closure of Drax, Britain's biggest coal-fired power station.

Yet the number of coal-fired plants is growing. The International Energy Agency (IEA), a think-tank funded by power-hungry countries, estimates that consumption of coal will increase by 71% between 2004 and 2030. Developing countries, in particular, rely on it. Coal provides some three-quarters of the power in both India and China.

The obvious solution is to make coal-fired generation cleaner. And that's what utilities in Western countries have been doing for years, to comply with ever stiffer air-pollution standards. Many literally wash coal to remove some of the impurities before burning it. Other technologies concentrate on purifying the smoke created during combustion. Small particles of ash, for example, are normally removed by forcing the flue gases, as the fumes from the furnace are known in the trade, between electrically charged plates. (Ash particles have a small electric charge, and are then trapped on one of the plates.) Other filters and chemical "scrubbers" catch oxides of sulphur and nitrogen that would otherwise cause acid rain.

Reducing emissions of carbon dioxide, however, is another matter. Most utilities tackle the problem indirectly, by attempting to improve the efficiency of their plants, and so to squeeze more electricity from each tonne of coal consumed or carbon dioxide produced. Most firms want to make such improvements anyway, since they cut costs and improve profits. In Britain, as in most rich countries, the average efficiency of coal-fired power stations is about 35%. But Mitsui Babcock, an engineering firm, says its most recent designs can achieve efficiencies as high as 46%. It reckons that switching from an old design to a new one can cut fuel consumption and emissions by 23%.

Most of the gains in efficiency come from increasing the heat and pressure of the steam used to turn a plant's turbines. The newest ones heat the steam to as much as 600 deg. C -- a state physicists call "supercritical". But there is no reason to stop there. Engineers believe that hotter boilers would raise yields even more. Such "ultra- supercritical" boilers, they say, could achieve efficiencies of over 50%, reducing emissions still further.

Substituting biomass for some of the coal burned can also help. Plants, after all, grow back after being harvested, so burning them does not add to overall levels of carbon dioxide in the atmosphere. By replacing 20% of their fuel with biomass, power stations can reduce their emissions by a further 20%. (Any more than that, says Lars Stromberg of Vattenfall, a European energy firm, and the ash from burning it would gum up the works of most furnaces.) Emissions also fall if biomass fuel is used to pre-heat the steam before it enters the boiler. All told, Mitsui Babcock calculates, these measures could cut emissions from coal-fired plants to the same level as those using natural gas.

The coal burnt in power plants can also be improved, mainly by drying. Less dense types, such as sub-bituminous and lignite coal, can contain up to 50% water. When burned, the water escapes as steam up the chimney, carrying valuable heat with it. Evergreen Energy, an American firm, is selling heat- and pressure-treated coal which, it claims, is as much as a third more efficient than ordinary coal.

Such techniques are particularly important in America, where power plants use a lot of sub-bituminous coal from the Powder River Basin in Wyoming and Montana. Coal from this region is low in sulphur, and so burns more cleanly, but has a relatively low energy density unless treated. Meanwhile, RWE, a German utility, is building a plant which will use the residual heat of its own flue gas to dry lignite before burning it. That increases the overall efficiency of the plant at little cost, since the coal can be treated without having to generate any extra heat.

These methods can reduce the various emissions produced by coal-fired power stations, so that they are at least no worse than gas-fired stations. But technologies also exist to make coal cleaner still, by filtering out carbon dioxide from the flue gas and storing it somehow. This is theoretically possible, but expensive. Other pollutants, after all, are essentially impurities, which can be washed from the coal or filtered out of the flue gas. But carbon dioxide is not a contaminant -- it is the inevitable by-product of carbon in the coal reacting with oxygen in the air. Along with nitrogen, the inert remainder of the air, carbon dioxide is the chief component of flue gas.

Gases, of course, are bulky and difficult to store. Most plans for carbon-dioxide storage involve liquefying it and pumping it underground into former oilfields, gasfields or coal beds -- an energy- intensive and thus expensive process. Separating the carbon dioxide from the nitrogen in the flue gas is also expensive, but necessary, since nitrogen turns liquid at a much lower temperature than carbon dioxide, and so requires even more energy to liquefy.

Moreover, unlike modifications that improve efficiency, there are no savings to be had by adding carbon-capture technology to a power plant. As a result, no such plants have been built. A few firms are building demonstration projects, while they wait to see whether governments impose long-term restrictions on carbon-dioxide emissions. Others, especially in Britain, which is blessed with natural storage tanks in the form of the declining oil and gasfields of the North Sea, have announced feasibility studies for clean-coal plants. Some utilities, such as AEP, are planning commercial plants that will initially lack carbon-capture facilities, but are designed to allow the technology to be added fairly easily at a later date. All are looking for government hand-outs or regulatory incentives to pursue these experiments, in the absence of any more compelling commercial logic.

How does carbon capture work?

Most utilities are eyeing one of three basic designs. The simplest, and easiest to bolt on to existing plants, treats carbon dioxide like any other pollutant, and extracts it from the flue gas. As this gas passes through a solution of chemicals called amines, the carbon dioxide is absorbed but the nitrogen is not. The carbon dioxide can later be released, by heating the solution, for subsequent liquefaction and storage. Many firms already use this "amine scrubbing" approach to remove carbon dioxide from natural gas, for example. But it is not so practical for large- scale uses, since the amines are expensive, as is heating them to release the captured carbon dioxide. The extra energy required would reduce a state-of-the- art supercritical plant's overall efficiency by about 10%, according to the IEA.

"Oxy-fuel" plants sidestep the difficulties of separating oxygen and nitrogen in the flue gas by burning coal in pure oxygen rather than air. The resulting flue gas is almost pure carbon dioxide. But the energy used to separate oxygen from air before burning is almost as great as that needed to filter out nitrogen afterwards, leading to a similar loss of efficiency. Oxy-fuel enthusiasts claim that modern plants can be easily retro-fitted to operate as oxy-fuel plants.

The third approach, called "integrated gasification combined cycle" (IGCC), also requires oxygen, but for use in a chemical reaction rather than for burning. When heated in oxygen, coal reacts to form carbon dioxide and hydrogen. An amine solution then absorbs the carbon dioxide, while the hydrogen is burnt in a modified furnace. The amine scrubbing is cheaper than usual, since the reaction generates carbon dioxide in a more concentrated form. Engineers are also experimenting with membranes that would allow hydrogen to pass, but not carbon dioxide.

There are four IGCC demonstration plants operating in America and Europe, although none currently captures carbon dioxide permanently; instead, it is simply released into the atmosphere. AEP's planned new plants will follow a similar design. The attraction of IGCC plants, aside from their carbon-capture potential, is that they produce fewer traditional pollutants and also generate hydrogen, which can either be put to industrial uses or burnt. But most utilities doubt that IGCC plants are suitable for mainstream power generation, because of higher capital costs and frequent breakdowns. Proponents retort that teething problems are natural, since IGCC is a newer technology, which combines previously unrelated processes from different industries.

George Bush is a believer, at any rate. In 2003 he unveiled a subsidised scheme to build a zero-emissions IGCC plant called "FutureGen" by 2013. The European Union, for its part, is giving money to utilities dabbling in oxy-fuel, among other schemes. Handouts from the taxpayer are needed, power firms argue, since the technology in question is still young. But it is hard to believe that it will ever grow up unless subsidies give way to stronger measures, such as long- term caps or taxes on carbon-dioxide emissions. The technology to eliminate such emissions from coal-fired plants exists, but it will not be adopted without regulatory incentives from governments.