After years of deliberation and extensive study, the next couple of months will finally see some definitive regulatory decisions on what should be done about mercury emissions from power stations in the United States, if anything. While this may not be exactly good news for operators of coal-fired plants, at least it may help to reduce some of the uncertainties. It also promises the possibility of interesting business opportunities – some have predicted a market potential of up to $5 billion – for those able to offer technologies that monitor and control mercury emissions.
The current flurry of interest in the difficult issue of mercury from power plants stems from the fact that the US Environmental Protection Agency has until 15 December 2000 to make a decision as to whether it intends to regulate mercury emissions from coal-fired power plants or not. If it does decide to regulate – and the conventional wisdom seems to be that it will – the Agency must then embark on a formal “rule-making” process and has a legally binding deadline of December 2003 to propose the regulation. A final regulation would be issued in December 2004, with full implementation by the end of 2007.
So the 15 December deadline is far from the end of the mercury emissions story in the United States, but could be seen as the end of the beginning.
This current phase of the US mercury emissions saga started with the 1990 Amendments to the Clean Air Act. These Amendments required the EPA to perform a study of the hazards to public health that might reasonably be expected to be posed by the remaining “hazardous air pollutants”, including mercury, emitted from power plants, following imposition of the requirements of the Clean Air Act. The EPA missed the original deadline for performing this study and for making a determination on whether to introduce further regulations on toxic emissions and was therefore sued by the National Resources Defense Council. In February 1998 the EPA did finally issue its report to Congress. This identified power plants as the largest remaining source of anthropogenic (ie caused by human activity) mercury, controls having been enacted for mercury emissions from other industries such as waste combustion.
But the Agency also concluded that it needed more information before it could decide whether to issue national mercury emissions standards for power plants. To force the EPA to address this information shortfall, an agreement was negotiated between the US government and the NRDC. This extended the EPA’s deadline provided it used its powers under the Clean Air Act to conduct a mercury “right-to-know” (RTK) survey. It was by this somewhat convoluted process that the 15 December deadline was arrived at.
The RTK survey has required the gathering of substantial quantities of information on the mercury content of power station coal, on mercury releases in stack gases, and on the potential for mercury reduction through controls on other emissions (SOx, NOx, CO2, particulates).
A mountain of relatively new data about power plant mercury emissions will therefore support the EPA determination, when it comes. It is all available on the web of course, or will be shortly, and presumably will be of great interest to other regulatory bodies around the world in formulating their policies on mercury.
The amounts of mercury involved – about 50 t per year for the entire population of US power plants – are minuscule relative to the total volumes of flue gas emitted. But the EPA position on the potential health implications of the mercury, in particular due to build up of toxic methylmercury in lakes and rivers, is supported by, among others, the National Academy of Sciences – which presumably can be regarded as reasonably objective in these matters.
As part of the information gathering exercise, throughout 1999 all coal-fired plants in the USA above 25 MWe – some 447 power plants, with about 1200 boilers in total – had to provide information on the mercury content of their coal – no mean exercise. Subsequently, stack gas monitoring was done on a subset of plants.
Currently the most mature technology specifically for mercury reduction seems to be the injection of dry sorbents such as activated carbon into the flue gas and collection of the sorbent by a conventional particulate control device such as an ESP or baghouse (fabric filters). Other potential dedicated mercury control technologies that have been or are being investigated include: carbon filter beds; condensing heat exchangers to extract residual heat from the flue gas; mercury capture using a noble metal (eg gold) sorbent; use of novel low cost sorbents such as those derived from biomass, tyres, zeolites and flyash; various types of catalysts; and even corona discharge processes, according to EPRI. Mercury control using these methods could prove difficult and expensive. One estimate has suggested that 90 per cent mercury reduction at a 500 MW plant might cost as much as $5 million per year. So it is just as well that systems designed to control other types of emissions (such as ESPs, fabric filters, FGD and SCR) can also be effective at reducing mercury. It is also worth remembering that the cost of mercury control technologies would fall should a market for them develop.
Nevertheless, with the growing price pressures on power plant operators in the liberalised market, the added burden of mercury controls, assuming EPA goes down that route, may be too much for some coal plant operators to bear, encouraging them to seek radical alternatives. One beneficiary might be the K-Fuel process, which, says its proponents, produces a coal that “contains 0.01 to 0.03 ppm mercury compared to 3 to 5 ppm mercury content of most other coals.”
