Old Gas, New Gas
Methane—made and taken apart by microbes, in the Earth, by people
At the wellhead, natural gas is 70 to 90 percent methane. The gas we
use at home—expensive, this winter—is nearly pure
methane. That simple, archetypical organic molecule, CH4,
methane, has a carbon bonded to four hydrogens at the corners of a
tetrahedron. And methane's structure is the only simple thing about it.
In a hurry to get to methane hydrates, I began by writing,
"Along with petroleum and coal, methane is a fossil fuel, of
plant origin…"—at which point I got corrected by
the president of Sigma Xi. When I changed what I wrote, geologists
gave me more trouble. I had stumbled right into a nest of
controversies. Or, an area of current research.
It appears that methane on Earth has not one source, but many. Most
(but not all) of the commercial methane in natural gas is
thermogenic—thought to derive from petroleum
(originally from plants) that is heated and processed deep
underground. It's old.
A great deal of additional methane, however, is sequestered in
sediments, at sea bottom and in permafrost, in a remarkable set of
structures I will soon describe. And its origins are controversial.
Much (some think all) is made by archaeans—the
neither-bacterial-nor-eukaryotic microorganisms that were only
distinguished in recent decades.
But there is an abiogenic source of sequestered methane too. Mantle
rocks that contain the mineral olivine (which describes a range of
minerals from Mg2SiO4 to
Fe2SiO4) are often altered to serpentine
change that also produces brucite ((Mg,Fe)(OH)2) and
magnetite (Fe3O4). The chemistry of this
"serpentinization" reaction is roughly this (to balance
the equation, we'd have to specify the olivine):
The important thing about this reaction is that the olivine is a
source of electrons that convert the protons in water to
H2. Combining it with CO2 results in methane,
courtesy of the so-called Fischer-Tropsch reaction:
This reaction proceeds in geological strata at accessible
temperatures and pressures in the presence of the necessary
catalysts. Some geologists think that most methane is created this way.
But the reaction also runs (catalyzed by enzymes now) in
microbes—methanogens—at temperatures hundreds of degrees
lower. To return to the very different setting of thermogenic
methane, geologist and chemist John M. Hayes of Woods Hole
Oceanographic Institution suggests that even there microbes could
have catalyzed thermogenic CH4 formation.
I am staying tuned. But let's return to that underwater methane,
wherever it comes from.
Under pressure and low temperature, methane (which normally boils at
-161 degrees Celsius) forms a thermodynamically stable association
with water. These solids are called methane hydrates, examples of a
broader class of structures, the clathrates.
What's stable at one temperature and pressure may not be at another.
Under ambient conditions at sea level, methane is a gas, water a
liquid. But in the permafrost and deep at sea, the weak hydrogen
bonds between water molecules reinforce the still weaker forces
between CH4 and H2O to create an aggregate
made of a water cage around one or more methane molecules.
Methane hydrates are white solids, less dense than water. They
remain on the seafloor only because they are agglomerated with rocks
and mud. (There, opportunistic evolution has led a variety of
species to use the methane in situ, as a carbon and energy
source.) Under the weight of 1,000 meters of ocean, methane hydrate
is stable to about 12 degrees, and because the seafloor is colder
than that, the ice-like hydrates form spontaneously wherever methane
is available. Brought up to the surface the hydrates fall apart to
methane and water.
In a previous issue of American Scientist (May-June 2001),
Robert L. Kleinberg and Peter G. Brewer looked at how gas hydrate
deposits might be exploited. Current guesses of the quantity of
methane contained in hydrates are around 1016 cubic
meters—exceeding by a factor of 100, roughly, our estimates of
"normal" natural gas resources. Could they be mined? Not
easily. Much of that methane hydrate is tied up in inaccessible
clays and pores. Moreover, one would have to get it out very
carefully, as methane is a most effective greenhouse gas. The
existing atmospheric burden of methane from natural-gas leaks, cows
and termites is consequential enough.
(An aside: In 2003, as part of its adherence to the Kyoto protocol,
the government of New Zealand proposed a flatulence tax on its 54
million sheep and cattle. New Zealand also has a population of
approximately 4 million people, but belching and farting ruminants
are responsible for roughly half of the country's greenhouse-gas
burden. The proposal was withdrawn after strong opposition from
farmers, but it still has supporters.)
One more lesson from the methane hydrates, part of my ongoing
struggle against the seductive forces of simplicity: The structure
at right, a dodecahedron of water molecules called 512
(referring to its twelve pentagonal faces), is only one building
block of the methane hydrates. The three most common hydrate
structures contain repeating units of 46, 136 and 34 water
molecules, some making up the 512 cavities. But the
common structures also incorporate polyhedra with four-membered and
six-membered hydrogen-bonded water rings of substantially greater
complexity, for instance 51268 and 435663.
A Kiln on a Landfill
A few years ago, I spent some time at the Penland School for Crafts
in the Blue Ridge country of North Carolina. There I met Jon
Ellenbogen, a potter with a good background in technology. Jon
showed me pictures of glass and ceramic kilns and of greenhouses on
the site of the nearby Mitchell/Yancey county landfill. Once a
smelly six-acre eyesore, the landfill now provides not only a home
for artists and greenhouses, but also, importantly, the energy they
consume. From methane.
The U.S. Environmental Protection Agency estimates that 33 percent
of methane released into the atmosphere by human activity comes from
landfills. This production is the result of microbes (they're at it,
around the globe) that decompose the buried waste. Landfill gas is
roughly 50 percent methane, 45 percent CO2 and a few
other gases. The total gas-generation potential of the
Mitchell/Yancey landfill is 2 cubic feet of methane per pound of
refuse buried. Figuring that each year the dump operated, it
swallowed perhaps 10,000 tons of waste, there's lots of methane
there. EnergyXchange, Ellenbogen's nonprofit organization, reasoned,
why not use it? If only more landfills around the country did so!
Biogas may be generated from any organic waste, but animal manure is
an especially good source. Large-scale animal productions of swine,
poultry or cattle are natural places to site biogas
digesters—dedicated residences for domesticated methanogens.
And there are already millions of small ones, adapted for household
use, installed around the world, especially in China and India.
The Indian biodigesters work mainly with cow manure—it takes
three good cows to provide a household with cooking gas and some
lights. The Chinese digesters, a somewhat different design, use
human nightsoil. In both designs, there is much useful digested
manure left in the end.
Meanwhile, Back on Titan
Just about a year ago, on January 14, 2005, the European Space
Agency's Huygens probe completed a one-way, 4-billion-kilometer
journey to Saturn's largest moon, Titan. The temperature there on
the surface is around -179 degrees, cold enough that methane is a
liquid. If you look at some of the startling images sent back
by Huygens you can see river-like channels that were probably carved
by streams of the stuff. Why methane? After nitrogen, it's the most
abundant component of Titan's atmosphere, and a heated sample of the
Titan soil released a puff of it.
So where does Titan's methane come from? People think it's that
serpentinization reaction mentioned earlier.
The C-H bond in methane is very strong. But there are bacteria,
methanotrophs, that have evolved to use methane as their carbon and
energy source. To do so, they must break the C-H bond of
CH4. This they accomplish with enzymatic finesse, using
methane monooxygenases, which contain a core of one or more copper atoms.
We now use petroleum as a carbon source for fuel, plastics, fibers
and pharmaceuticals. That resource will soon be exhausted. Methane
will be around longer. One idea is that we might farm those bacteria
to give us carbon feedstocks from methane. But could we invent an
efficient industrial process that breaks the strong C-H bond?
Perhaps the nitrogen fixation story provides a lesson here. While
our biochemistry craves nitrogen atoms, we cannot fix abundant
atmospheric nitrogen. Some bacteria can; biologically assimilable
nitrogen also comes from minerals in the soil and naturally acid rain.
But the natural sources do not suffice. Fritz Haber and Carl Bosch
devised (over ninety years ago) an industrial process to make
ammonia from N2 and H2. The chemistry is so
successful, so economical, that today over half the nitrogen
atoms in our bodies have seen the inside of a Haber-Bosch
factory. If all those factories disappeared, there would be
enough fixed N for only half the people on earth.
To put it another way: In providing us with a key element, N,
cultural evolution (science and technology) competes effectively
Returning to methane: Of course, there is one reaction we all
know—burning—which certainly activates methane. The
problem is that it does it too well, taking CH4 all the
way to inert CO2. Along the way, inside hydrocarbon
flames, are the partially oxidized "intermediates" that
industry wants. But the process of burning is nonselective; it does
not stop, for example, at methanol, CH3OH, a molecule we
In principle, several desirable reactions are feasible, such as the
ones that lead methane to methanol or acetic acid, or produce, using
methane, ammonia and hydrogen peroxide. Indeed, several commercial
processes currently begin with methane and convert it to such
molecules. The problem is that they require really high
temperatures. For selective, low-temperature chemistry, a catalyst
Alexander E. Shilov of the Russian Academy of Sciences came up with
the first candidate in 1969. Using platinum salts, he saw
hydrogen/deuterium exchange in alkanes (compounds of just C and H,
such as methane) that had been mixed with deuterated water,
D2O. Therefore, some of the C-H bonds must have been
broken and reforged—the first evidence that the C-H bond was
Passing over, unwillingly, much beautiful chemistry, we come to the
exciting, recent work by Roy Periana and his coworkers at the
University of Southern California, who have improved on Shilov-like
chemistry to the stage that they can convert methane to methanol
with a yield of better than 70 percent at 220 degrees. They can also
convert methane directly to acetic acid at 180 degrees—much
cooler than the >800-degree conditions of other processes.
Periana thinks that within 10 years (I guess 50) we will have all
the carbon-containing molecules we need—and many we haven't
thought of—made from coal and methane sources. Goodbye petroleum.
© Roald Hoffmann
Thanks to Martin Fisher, John Hayes, Martin Hovland, Jay
Labinger, Lynn Margulis, Roy Periana and Norman R. Scott for
their enlightening comments.