The Economist

August 5, 2000

The Dawn of Micropower

micropower links

THOMAS EDISON was a man of great foresight, but
who would have thought he could have been more than
100 years ahead of his time? When he set up his first
heat-and-electricity plant near Wall Street in 1882, he
imagined a world of micropower. Edison thought the best
way to meet customers’ needs would be with networks of
nimble, decentralised power plants in or near homes and
offices. What goes around, comes around. After a century
that seemed to prove Edison wrong—with power stations
getting ever bigger, and the transmission grids needed to
distribute their product ranging ever wider—local
generation for local consumption is back in fashion.

There are several reasons for this. One is market
liberalisation. About half of America’s state governments
have now forced their erstwhile electricity monopolies to
face competition. In the European Union, a directive that
took effect in 1999 ordered member governments to open
up part of their wholesale market for electricity. Many
developing countries, too, from India to Argentina, have
embraced deregulation and privatisation.

Small, local power plants offer a cheap way into such
markets. Even if the power they produce is more costly at
source—which it often is—they do not suffer huge
transmission losses when sending it to consumers. On top
of that, the surplus heat they generate can be employed for
useful purposes, such as warming buildings, whereas that
from big generators located in the middle of the countryside
is usually wasted. The result is that local power generation
has now become economically competitive.

A second reason for the rise of micropower is
environmentalism. Ever-higher emission standards have
made it unattractive to build new coal-fired plants in the
rich world. America still gets more than half of its electricity
from coal, but only because many older plants have been
“grandfathered”, and so do not have to meet strict new
emissions standards—a derogation that is almost certain to
be struck down at some point. Europe has been even more
aggressive than America in pushing industry to adopt
cleaner forms of power generation. And microgenerators
are exceedingly clean. The worst of them burn natural
gas—a reasonably benign fuel. Others use hydrogen and
sunlight, both environmentalists’ dreams.

A third, increasingly important reason is the demand for
reliable, uninterrupted power. Karl Stahlkoph, the head of
the Electric Power Research Institute (EPRI), an
industry-financed American research body, reckons that
micropower will take off in America, where brownouts and
blackouts are an ever-increasing problem, as much
because it is under its owners’ control as for any other
reason.

These three things have stimulated the search for small,
clean, reliable and above all cheap generating technologies.
And such technologies are now emerging, fuelled by a
surge in venture-capital investment (see chart) and the
prediction that, within a decade, the market for such
equipment may be more than $60 billion a year.

Powerful choices

The most dramatic breakthroughs are taking place in the
field of fuel cells. These devices, which work by combining
hydrogen with oxygen from the air to produce electricity,
are popular candidates to replace internal-combustion
engines in road vehicles. But they look increasingly
plausible as replacements for power stations, too.

There are several sorts of fuel cell, but all consist of two
electrodes (an anode and a cathode) separated by a
material called an electrolyte. In most fuel cells hydrogen is
fed to the anode, where it is ionised into a proton and an
electron. The proton makes its way to the cathode through
the electrolyte, while the electron goes there the long way
round—via a wire that leads into whatever the fuel cell is
powering, and back again. At the cathode, the protons and
the electrons react with oxygen from the air to make water
which, to the joy of environmentalists, is the only waste
product of such a cell.

The leading fuel-cell technology at the moment is generally
reckoned to be the proton-exchange membrane (PEM) cell.
In this, the electrolyte is a polymer membrane coated with
platinum, a metal that acts as a catalyst for the chemistry
involved.

Ballard Power Systems, a Canadian firm, is the leading
proponent of PEM technology. Firoz Rasul, its boss, says
he expects his firm’s first commercial product to reach the
market next year. This will be a 1kW generator, to be
marketed by Coleman, an American outdoor-goods firm,
for household use. Ballard is also developing a power unit
with Tokyo Gas, a utility that supplies Japanese homes with
natural gas. That version would “reform” the natural gas
first, by reacting it with steam to release the hydrogen in it.
This means the exhaust will include carbon dioxide. But
reformation eliminates the need to supply the cell with pure
hydrogen, making the whole process cheaper.

A rival to PEMs is the solid-oxide fuel cell. A leading SOFC
design arranges an electrolyte and two electrode layers in a
tube. Air flows through the inside of this cell and hydrogen
past the outside. In this case it is the oxygen that is ionised
(by heating the air up to 1,000°C), and thus supplies the
electrons. Although SOFC units have to operate at higher
temperatures than PEM cells, they can achieve levels of
efficiency much greater than is now possible with PEMs.

Siemens Westinghouse, a big power-equipment firm,
expects to bring SOFCs to market in 2004, at a price of
$1,500 per kW, dropping quickly to the $1,000 threshold
that is currently achieved by coal-fired power stations.
And, unlike Ballard with its 1kW units, Siemens is building
generators capable of producing between 0.3MW and
10MW. These are aimed at industrial customers.

A third variation on the fuel-cell theme is the alkaline fuel
cell. This requires two porous electrodes, separated by an
electrolyte composed of potassium hydroxide. ZeTek
Power, a British firm that is due to go public early next
year, is leading the development of this technology.
Nicholas Abson, the firm’s chief executive, insists that his
technology is cheaper, easier to make and more practical
than either SOFC or PEM cells. Unlike SOFC, alkaline cells
work at relatively low temperatures. Unlike PEM cells, they
do not rely on platinum catalysts. ZeTek, according to Mr
Abson, has perfected the use of cheap metal-oxide
catalysts that will help to bring the cost of its stationary
fuel-cell systems below $500 per kW within 18 months.

Less is more

Fuel cells are a nifty idea, but they suffer from one serious
disadvantage: that the world is not set up to deliver
hydrogen cheaply. Technologists are working on this
problem. Hydrogen for Ballard’s cells is stored in
substances called metal hydrides, which can absorb large
quantities of the gas. But systems that can make use of
existing fuel-delivery infrastructures are likely to have a
head start—as Ballard has conceded in its deal with Tokyo
Gas.

A second novel micropower technology, however, is
ideally suited to natural gas. This is the microturbine. The
clever thing about a microturbine—as opposed to the big,
clunky sort of turbine that is used in traditional power
stations—is that it has only one moving part. This is a
high-speed compressor-cum-rotor that spins at up to
100,000 revolutions a minute.

The near-absence of moving parts means that
microturbines are cheap to operate and maintain—costing
as little as a third of the running costs of a comparable
diesel generator. Even the problem of lubricating the one
part that does move seems to have been solved. Capstone
Turbine, a small American firm, has developed a version of
the device that uses sophisticated “air bearings” which
require no liquid lubrication. Capstone, unlike many other
companies in the microturbine market, is already selling its
products—shipping several thousand a year, ranging in size
from 25kW up to 500kW, to a number of commercial
clients.

The third aspirant micropower technology is solar energy.
Like fuel cells, which were first dreamed up in the 1830s,
photovoltaic solar cells have been a long time coming as an
everyday means of power generation. But they are almost
there.

Solar cells are composed of a semiconductor such as
silicon. When the sun’s rays hit a cell’s surface, some of the
semiconductor’s electrons absorb enough energy to rush
off towards the other side of the cell, where a lattice of
delicate wires embedded in the surface gathers them up
and feeds them into a cable.

The advantages of small solar-power plants are that they
are clean, reliable and, of course, that the fuel comes free.
The snag, however, is that the equipment does not. The
energy from such plants costs between 22 cents and 36
cents per kW-hour, twice the expected cost for fuel cells.

Those costs, however, are a quarter of their level two
decades ago, and look likely to fall further thanks to
breakthroughs in the manufacture of the silicon wafers from
which solar cells are cut. AstroPower, the only integrated
solar-energy firm to be traded publicly, has come up with a
very-high-speed manufacturing process which it calls
“silicon-film” making, and which is akin to the “float glass”
method used to make window panes. This should halve the
cost of wafers, bringing the technology’s price within
spitting distance of its rivals.

Back to the future?

The new micropower technology is undeniably impressive.
But the big question is whether the market for distributed
generation will take off this time—over a century after its
first bloom. One reason to think it might is that its costs
have come down to economic levels (see chart). The
trends suggest they will fall still further over the coming
decade, making micropower attractive to the ordinary
consumer in the rich world.

The greatest potential for micropower, however, may lie in
helping the 3 billion people in the poor world who have no
reliable access to electricity. Gary Mittleman, the boss of
Plug Power (a firm which, in collaboration with GE, a big
American electrical company, is one of Ballard’s rivals in
the PEM market), reckons that it costs between $1,000 and
$1,500 per kW to build or replace electricity grids in
developing countries. In such places, micropower is
already an attractive option. International agencies such as
the World Bank, as well as private-sector operators and
non-governmental groups, are devising “microfinance”
schemes to help bring electricity to the poor in such
countries as Mongolia and India.

In time, micropower may also change the way electricity
grids themselves operate—turning them from dictatorial
monopolies into democratic marketplaces. Add a bit of
information technology to a microgenerator and it will be
able both to monitor itself and to talk to other plants on the
grid. Visionaries see a future in which dozens, even
hundreds, of disparate micropower units are linked
together in so-called “microgrids”. These networks could
be made up of all sorts of power units, from solar cells to
microturbines to fuel cells, depending on the needs of
individual users. EPRI has feasibility studies under way to
develop a microprocessor-based converter that will enable
“plug and play” connection of any micropower device to
the power grid.

As energy markets liberalise, online energy-trading spot
markets develop, and individual consumers win the right to
select their energy suppliers, some even see the emergence
of “virtual utilities”. Microgrids would allow such firms to
combine the individual efficiency of micropower plants with
the market power that is gained by bundling together their
collective generating capacity. Whether run in competition
with established utilities, or by them, such virtual utilities
would, according to Goran Lindahl, head of ABB, a large
European generating-equipment company, result in “greater
system reliability, lower operating costs, reduced
environmental impact and improved overall business.” ABB
is now building microgrids that should be up and running by
2001 in both Europe and America.

Much as with the Internet, the companies that develop the
technology to allow the electricity grid to perform intelligent
metering and switching, and that position themselves as
“air-traffic controllers” for these streams of electrons, will
lead the industry. It is a heady vision for what many think of
as a dull commodity business. Edison would surely be
proud of the role that micropower looks likely to play in the
third century of the electricity age.

LINKS
Click to visit the companies and organisations
mentioned in the article: EPRI, Nth Power, Ballard
Power Systems and Capstone Turbine. Click the terms
for a diagram and short description of solid oxide fuel
cells and proton exchange membrane fuel cells, or read
more about microturbines. ZeTek Power has a page
explaining its alkaline fuel cell technology, while Power
Plug has a page with good links and a basic overview of
fuel cells. Seth Dunn’s report “Micropower: The
Next Electrical Era” from the Worldwatch Institute, or
listen to the author speak about micropower.
Worldwatch also keeps an excellent page of
micropower links.