Its 'common knowledge' that there is 250 - 300 years energy supply locked up in coal. In this case common knowledge is wrong. We use a lot more energy these days compared with the rate it was being used when that forecast was made. There's a hidden assumption too: that energy use will remain at about the level that applied when that estimate was made. When Einstein was asked what the greatest known force was he said "compound interest", so lets apply that. If we assume there is a 250 year coal supply at current consumption rates and coal use increases by 2% a year it will all be gone in 85 years. Convert that coal to synthetic petrol, a process that is only 60% energy efficient, and its all gone in 50 years.
A recent review (New Scientist, vol 197, no 2639, 19 Jan 2008 page 38) of coal reserves supports my simplistic analysis. It makes the point that the industry accepts its estimates to be quite inaccurate and more than somewhat inflated. A practical indication that we may be exhausting the supply of good quality coal is that the price quintupled between 2002 and 2008 but the supply hasn't increased as orthodox economic theory says it should. David Rutledge, an engineering professor at Caltech, used Hubbert Linearisation, which extrapolates total stocks by plotting extraction rate against the total amount produced, to estimate coal stocks. His trial run using UK coal figures showed that UK coal mining peaked in 1913 and its decline since was due at least as much to working out the mines as to technology changes and different choices of fuel. He has estimated world coal reserves at 450 billion tonnes, which is slightly more than half the industry estimate. Combining that with forecast demand suggests that we'll pass peak coal not more than a decade or two after peak oil: coal is unlikely to become an oil substitute.
This discussion ignores the ecological damage resulting from carbon dioxide and the other combustion products of coal. If the Hubbert Linearisation estimate is right, atmospheric carbon dioxide may just exceed the runaway greenhouse tipping point of 480 ppm. In other words, burning all the available coal while emiting its combustion products into the atmosphere is not a good idea.
It seems that coal is not exactly a long-term energy resource.
Coal-powered electricity generation is the biggest single source of carbon dioxide in the UK. Currently its just emitted to warm up the climate. In some quarters a popular solution to climate change is carbon sequestration.
In addition, coal contains significant amounts of sulphur, which is burnt to sulphur dioxide and trioxide. If these are emitted, they dissolve in rain to form sulphuric acid. Hence the need to extract sulphur oxides from the exhaust gases to prevent acid rain from harming forests and acidifying lakes. Sulphur removal produces calcium sulphate, better known as Gypsum. Some is used by the building trade, but there's still a lot left over to dump.
On top of all this there's a final snag. Generation efficiency rises as the superheated steam temperature rises, which in turn needs a higher combustion temperature. Unfortunately a side effect of this is that atmospheric nitrogen burns to form nitrogen oxides, another damaging atmospheric pollutant. Once again, more expensive exhaust cleaning is needed, but at least the by-products of this are saleable.
The Germans, during World War 2, and South Africans, during the Apartheid sanctions, were forced to make petrol from coal but the result was expensive and inefficient. Its a very energy-intensive process. 40% of every tonne of coal converted to petrol is burnt to provide energy for the conversion and only 60% ends up in the petrol storage tank. Think "lots of carbon dioxide" again.