Natural gas is bad for the environment!
What is natural gas?
What happens if you burn methane?
What happens if you don't burn methane?
What is the composition of NG?
Purification of NG
Distribution of NG
Liquefaction
Compression
Pipelines
Emissions
Calculation of total climate change effect
Conclusion
References
There has been controversy in Cyprus as to the use of natural gas to
generate electricity and whether it should be imported as liquid (safe but
expensive) or compressed (less safe but easier to use). This essay is aimed
at showing that neither option is ideal.
The producers of natural gas claim that it is the most "ecological" of
all fossil fuels. If examined as fuel at the point of use, this may be true.
However, this is very untrue if studied holistically and is, probably, the
worst of all forms of fossil fuels. The purpose of this essay is to
demonstrate that natural gas is a very heavy contributor to
greenhouse gas emissions in many ways,
probably even worse than coal, and its use should be curtailed to a minimum.
Natural gas (NG) is produced by the anaerobic decomposition of living
matter. It was so named because one can see bubbles of it rising from the
sediment of stagnant pools or the "will-o'-the-wisp" over marshland.
Chemically, it is methane. It can also be collected from composting.
However, probably more than 99% of the so-called natural gas that is used
throughout the world as a fuel is anything but natural today, although its
origin may have been 100 million years ago. It is extracted by man from deep
in the earth's crust. As such, it is a fossil fuel, just as much as coal or
oil. It is a fossil fuel which is 75% carbon.
Theoretically, whether it powers a fuel cell, is burnt in a gas ring, in
an internal combustion engine, in a gas turbine, in a power station or in a
central heating boiler, the result is the same. It reacts with oxygen in the
air to form carbon dioxide and water vapour:
CH4
+ 2O2 > CO2 + 2H2O
This reaction is very exothermic, that is, it produces a large amount of
heat. In other words, the chemical energy stored in the methane is converted
into thermal energy that may be used to cook a pan of spaghetti, drive a car
or generate electricity etc.
Assuming the combustion is complete, the only apparent pollutant produced
is the carbon dioxide, but, for every kilogram of methane that is burnt,
2.74 kg of carbon dioxide is produced, yes, nearly 2¾ times as much. Carbon
dioxide is the most prevalent greenhouse gas, responsible for
climate change.
Unfortunately, it is not as simple as that. There are four factors which
make matters worse:
- combustion is rarely complete without some means of post-combustion,
so some methane escapes to the atmosphere.
- this scenario takes no account of what has happened before the
methane reaches the consumer
- the combustion also produces NOx gases which are
precursors to photochemical smog and
tropospheric ozone, responsible for
much pulmonary disease
- commercial natural gas is only 95 - 99 per cent methane; the
remaining compounds may include small quantities of sulfur and
radionuclide gases (radon) which are responsible for more serious
pollution after combustion.
This where a large part of the crunch lies. Methane in the atmosphere is
a powerful greenhouse gas. In fact, it is between 20 and 50 times worse than
carbon dioxide, depending on what are called the free hydroxyl radicals that
are present in the atmosphere, which are very variable (low concentrations
of OH radicals decompose methane more slowly than high concentrations). Let
us assume, for ease of argument, that the global average
Global Warming Potential is 35, although it
is likely to be higher in desert country and in polar regions, both with
considerable NG production. (I have noted that some web sites authored by
vested interests in NG cite the figure of 21, a figure which would be
typical of a hot, very humid climate such as in a tropical rain forest. This
low figure is not realistic in real life, except in a few relatively minor
producing countries, such as Malaysia, Indonesia and Brunei.)
It is therefore clear that allowing NG to be emitted is far worse for
climate change than burning it.
Natural gas, as it comes out of the ground, is very variable in
composition, depending on the location. An average composition, synthesised
from many sources throughout the world, and which I'll use for further
discussion, is:
| Component |
Percentage |
| Methane |
85 |
| Ethane |
8 |
| Butane |
1 |
| Propane |
0.5 |
| Heavier HCs |
0.1 |
| Nitrogen |
1 |
| Carbon dioxide |
2 |
| Hydrogen |
0.1 |
| Oxygen |
0.1 |
| Hydrogen sulfide |
0.5 |
| Water vapour |
1.2 |
| Other gases |
0.5 |
NG comes from both ad hoc gas wells, but also as a boil-off from
crude oil (bubbling, much like the gas in soda water), the latter with more
heavier HCs.
As I mentioned earlier, commercial NG is usually 95 to 99 per cent
methane, averaging about 98 percent, the other 2 per cent being mostly
ethane with traces of all the other gases. This implies that the NG must be
purified before it is distributed.
There are many processes used for NG purification. A typical process line
may include:
- removal of oil and condensates by cooling and settling. Some of
these may be further purified for commercial purposes.
- removal of water by absorption in diethylene glycol in a tower
followed by adsorption in
zeolites
- removal of propane and butane by absorption and fractional
distillation. These are of commercial value as bottled LPG.
- removal of ethane by cryogenic techniques. After distillation, this
is useful in the petrochemical industry.
- removal of sulfurous gases and carbon dioxide by absorption in
monoethanolamine. In cases where the
sulfur content is high, it may be economically viable to separate it.
At each of these stages, there is a small concomitant methane loss,
mainly due to recycling the diethylene glycol and monoethanolamine and
reactivating the
absorbants and
adsorbants.
A purification plant is an important infrastructure and one plant may
serve many wells over a considerable area, with a spider's web of small bore
pipework. This is often cast iron pipes with flanged joints, notable for
leaks. At the wellhead, there is a "tree" for initial separation of gross
impurities, including sand, by purging them out with the gas.
The unpurified gas may be very corrosive, especially from "sour" gas
wells with high water vapour and sulfur content. This means the lifetime of
the pipework to the purification plant is limited and it must be regularly
inspected for leaks.
Once purified, the gas has to be distributed to the end user. This is
done by either liquefaction (LNG) or compressed by pipeline (CNG). Of
course, the liquefied gas is eventually returned to gaseous state and
compressed.
LNG is produced by refrigeration down to -163 °C at atmospheric pressure.
It is then stored in large, double walled, well-insulated, spherical or
cylindrical tanks in high-nickel steel, rather like enormous Thermos flasks.
These tanks are not pressure vessels and have to be vented by pressure
relief valves at, typically, 300 hPa, so that there is no risk of damage as
the contents heat up, no matter how good the insulation.
The liquefaction process itself is done in two stages, initially a
pre-cooling in a propane refrigeration circuit and then in a mixed gas one.
It is quite a complex process requiring a great deal of energy. This is
often supplied by gas turbines using the gas vented from the storage tanks
at the liquefaction plant and raw methane. The resultant liquid-phase
methane has a volumetric ratio of 1:593 compared to gas-phase methane.
Unfortunately, I have not been able to obtain figures for the emissions or
gas consumption (energy) at liquefaction plants but they are far from
negligible.
Methane gas is easily converted to CNG. Three kinds of energy source for
compressors are used: gas turbines running on NG, reciprocating engines
running on NG and electric motors. Pressures up to 240 bars are sometimes
used for bottled methane, but most pipelines run at 15 to 100 bars.
Some leakage is almost inevitable with compressors, especially as they
age. Much maintenance is required to keep emissions to a minimum.
Most major pipelines are constructed of rolled sheet pipes with a welded
seam and with sections welded together. When new, these are almost perfectly
leak-free. However, they are generally buried at depths of typically 1.5 - 3
metres and the steel can corrode either from within or without. Corrosion is
minimised by treatment with a coal tar coating, but this does not last for
ever and leaks do develop over time.
Compressors are placed every 50 - 150 km along a pipeline, and isolating
valves at 10 - 30 km intervals. Small leakages occur at every valve, through
the stuffing and flange gaskets.
Pipelines need to regularly checked and maintained. Leaks are detected by
portable gas detectors along the ground over the pipes. However, the
greatest emissions are made when "pigging" a pipeline. A section of pipeline
is isolated by closing the valves at each end and unscrewing the flanges. A
very high-tech robot, nicknamed a "pig" is introduced into the pipe and sent
from one end of the section to the other, examining the internal surface for
weld problems, corrosion or leaks and transmitting the information back to
an analytical computer. Obviously, this vents the gas in the section. After
"pigging", the section has to be purged of air, before it can be put back in
service and this, too, also involves considerable emissions. It is rare to
either collect or flare the gas in the sections.
Small pipelines, particularly distribution pipelines in cities, are often
relatively small bore flanged cast iron pipes, especially in older quarters.
The leaks at the flanges are often aggravated by vibration from heavy
traffic. Newer ones may be extruded steel from a punched blank, with welded
joints, but street stop cocks are inevitably flanged. Some household
distribution systems use welded plastic pipes, which are inevitably slightly
porous.
It is difficult to obtain precise figures of emissions. Global estimates
vary between 25 - 70 teragrams where 1 Tg = 1012 g or 1 million
tonnes. This represents about 3 - 9 percent of all NG extraction. For
convenience, I'll assume an average of 5%, although the precise figure
cannot be substantiated. This represents about 40 Tg of methane, which will
have the same climate change effect as about 1,400 million tonnes of carbon
dioxide, a far from negligible quantity. The figure of 40 Tg is probably
conservative as one major source cites 45 Tg (Tetlow-Smith, 1995). For
comparison, the total weight of carbon added to the atmosphere annually from
the combustion of fossil fuels is estimated at 7,000 million tonnes, of
which 2,000 Mtonnes are naturally sequestered, mainly in the oceans.
The emissions can be divided into those:
- produced during drilling a well, up to the moment of capping
- fugitive emissions due to equipment leaks
- fugitive emissions due to pipeline leaks
- vented leaks from pressure relief
- vented leaks for maintenance
- vented leaks from diethylene glycol, monoethanolamine and adsorber
regeneration
- due to incomplete combustion of distributed NG
The biggest studies of emissions have been carried out in the USA, with
some 6 - 7 Tg/year, representing about 1.4 - 2 percent of NG consumption.
However, this figure excludes unburnt gas emissions at the users' premises,
drilling emissions and those due to gas produced and transported outside the
USA but consumed in the country. As imports of LNG were over 15 percent of
all NG consumption in the USA in 2002 and are expected to top over 20
percent in 2005, this is not a negligible factor considering that shipping
LNG over several days to a few weeks from liquefaction to regasification
will involve considerable emissions. It is therefore expected that the
holistic figure of emissions due to all gas consumed in the USA will be as
high as 4 to 5 percent. In many other countries, especially those with poor
equipment and antiquated pipelines, the percentage of emissions would be
worse.
Domestic emissions in the USA, excluding well-drilling and
extraterritorial emissions of gas consumed in the USA and also excluding
emissions due to incomplete combustion at users' premises, have been given
as follows:
| Source |
Percentage of total emissions |
| Fugitive: from compressors |
21.5 |
| Fugitive: production facilities |
5.5 |
| Fugitive: gas plants |
7.8 |
| Fugitive: metering and pressure regulating |
10.1 |
| Fugitive: users' meters |
1.8 |
| Fugitive: underground pipelines |
15.4 |
| Vented: pneumatics |
14.6 |
| Vented: maintenance purging |
9.6 |
| Vented: chemical regeneration |
4.0 |
| Vented: dehydrator |
1.5 |
| Vented: other |
0.3 |
| Combusted: compressor exhaust |
7.9 |
From the above data, an approximation of the effect that the use of NG
will have in the climate change equation may be made (figures rounded off):
Assume that the user consumes, in a given length of time, 100 kg of NG,
of which 98 percent is combusted. This will produce 98 x 2.74 = 268 kg CO2
+ 2 kg methane = 70 kg equivalent CO2 = 338 total
Assume that 5 kg of methane has been emitted from the well-head to
consumer system = 175 kg eq. CO2 = 513 cumulative total.
Assume that the energy requirement for the transport (compressor,
liquefaction etc.) has consumed 5 kg of NG, totally combusted = 14 kg CO2=
527 cumulative total
Assume that 110 kg of gas is required from the purification plant
and the input gas from the well-head is 85% methane and the conversion
efficiency of the purification plant is 90%. 144 kg of gas is required for
the process, of which 14 kg is converted to CO2 from the process
= 38 kg CO2 = 565 cumulative total
Assume 144 kg of gas produces 11 kg of ethane, 1.5 kg of butane and 1 kg
of propane, which are all subsequently converted to CO2 (burnt or
decomposed), producing respectively 16, 1 and 1 kg CO2 = 583
cumulative total of equivalent CO2
Assume 144 kg of gas produces 3 kg of CO2 = 586 cumulative
total of equivalent CO2
For comparison, burning 100 kg of pure carbon would produce 367 kg of CO2
Best Welsh anthracite coal is 91 percent carbon and 7 percent hydrocarbons.
It would need about 115 kg of anthracite to equal 100 kg of methane in terms
of usable heat produced (the equivalence is difficult to calculate as the
difference in temperature of combustion makes losses non-equivalent). This
would produce about 445 kg of CO2 or about 25 percent less than
methane. However, this comparison is not strictly fair, because the carbon
dioxide produced during the mining and transport of the anthracite has not
been calculated in. Nevertheless, even if we add an extra, say, 15 percent
for this, natural gas produces more greenhouse gas than coal when viewed
holistically.
Natural gas is not the least polluting of fossil fuels, as the large oil
producers would have us believe. In terms of purity, it is good, but
greenhouse gas emissions are holistically very high from its use. It has
been proposed as a substitute for petrol in internal combustion engines, but
it is believed that this will increase greenhouse gases, especially as the
disconnection of pressure hoses at filling stations will inevitably release
raw methane into the atmosphere.
Many approximations have been made in these calculations, but these have
been made conservatively and in good faith. Unfortunately, accurate data
permitting a better calculation are not available.
I conclude that the use of natural gas would be better curtailed if we
are to improve our record for greenhouse gas emissions, something that is
particularly important in the Cyprus context because of the fact that
virtually all our energy is carbon-emissive. As a final word, methane
concentrations in the atmosphere have increased to 2.25 times the
pre-industrial level, compared to only 1.3 times for carbon dioxide,
entirely due to man-made causes. As the atmospheric residency time of
methane is only a small fraction that of carbon dioxide, even with low
hydroxyl radical concentration, cutting emissions would have a much faster
effect on reducing climate change effects than cutting down on other fossil
fuel combustion.
Tetlow-Smith, A. (1995) Environmental factors affecting global
atmospheric methane concentrations. Prog. Phys. Geog. 19,
336-50
|