The cryptic terms “V2G” and “V2H” have recently appeared in the vocabulary
of those who advocate the widespread adoption of solar and wind power. By
“vehicle-to-grid” and “-home” respectively, they mean to suggest that a
million electric cars sold would represent something like twenty
gigawatt-hours of storage batteries available to compensate for the irregular
supply of energy from their favoured sources, without appearing on anyone’s
power bill.
The principal function of an automobile is driving. Any discharge from the
battery for some other purpose reduces the range which can be driven before
stopping to recharge. This by itself should give us reason to doubt that a
car battery is a likely place to stash excess power, either to feed back out
to the distribution network when demand exceeds supply, or to meet domestic
needs overnight.
Then there is the question of just how an automobile is used. Most people
do most of their driving during the day, and their cars are most likely to be
away from home, either on the road or in a parking lot, during the peak hours
for solar generation (approximately 10 AM to 3 PM). Accordingly, charging is
mostly done overnight at home, a pattern reinforced by the fact that few
parking lots now have car charging stations, and those which do have perhaps
one or two.
All this assumes that electric vehicles are used in the same way as their
gasoline analogues, which is reasonable at least for the first few years. As
market penetration increases, however, and the electric car becomes less of a
specialty item, economic considerations will begin to make themselves felt.
The median price of a new private automobile in the United States is now above
thirty thousand dollars. Any businessman who had purchased a piece of
equipment at such a price, only to have it stand idle twenty hours or more out
of the twenty-four, would think himself to have made a poor investment.
The cost of operating an electric automobile bears a much smaller proportion
to its prime cost than is true for a gasoline automobile. This creates a
considerable economic incentive to achieve high utilization, perhaps by taking
advantage of the various short-term car-hire and similar services now in the
market. Hence we may expect that, as charging stations become more common, a
substantial fraction of cars will linger only just long enough to ‘top off’
before setting off on another trip. The converse case, of someone who owns
two electric cars and leaves one to power his house, while the other soaks up
cheap midday power at some other location, we may expect to be reasonably
rare.
Hence, even without examining the fundamental point that a battery pack
cannot be cheaper with a car wrapped around it (about one dollar per watt-hour
at the best present-day prices) than without, money which comes out of the
consumer’s pocket one way or another, we are entitled to expect that
battery-electric automobiles will make little if any substantial contribution
to matching general electricity demand to the supply from intermittent
sources.
“Sustainable” is a word which one hears bandied about. What does it mean?
Heraclitus could have told you that the answer to the question “can we keep
doing the same thing indefinitely?” is no. The world is
always changing, and human activities must change with it. Indeed, at least
since the inventions of gunpowder and the steam engine, perhaps since the
invention of agriculture, human activity has become one of the important
sources of change in the world we inhabit.
Perhaps a reasonable definition is that, in the context of a technological
civilization, a sustainable approach to any problem is one which, of itself,
imposes no hard limits on dependent developments. So, since energy is a basic
factor of production, a sustainable source must provide an annual supply large
enough to meet the needs of a reasonably foreseeable global population with a
reasonably high standard of living, at a cost low enough to make that viable,
without giving out in the foreseeable future. We should not be too far wrong
in looking for three times the current global primary energy production, at a
retail price to domestic customers of five cents per kilowatt-hour (equivalent
present-day value), for a minimum of one thousand years.
The quantity of coal laid down in the crust of the Earth during the
Carboniferous Period is immense, but the rate at which it is being dug out is
likewise great. This drives mining toward ever less accessible, ever more
costly deposits. Long before the total quantity can be exhausted, the cost of
recovery will have risen to the point that coal will not be able to supply
power at prices low enough to support general use. With petroleum, which
(although down from its peak of almost half the whole world energy supply) now
accounts for somewhat more than coal, and natural gas, which is following
close behind, that point is
already in sight. The exploitation of tight shales, deep and remote
seafloor deposits, and tar sands is already accepted practice, and while
technological advances have helped to restrain prices, ultimately the energy
required for extraction will exceed that available from burning the fuel. If
we account for the external costs of using fossil fuels, the point of
uneconomy must come much sooner.
The phrase “too cheap to meter” has sometimes been used to describe atomic
power, and is generally misunderstood. Its import is not that the absolute
cost of a kilowatt-hour from nuclear generation is exceptionally small
(although that figure has proven moderate), but that the marginal
cost per kilowatt-hour is not large in comparison to the fixed charges
associated with generating and distribution plant. Accordingly, it might be
reasonable to charge each subscriber a fixed monthly fee based on the size of
his service connection, perhaps with a surcharge reflecting the total load on
the system, rather than go to the expense of accounting for the actual power
consumed. (The surcharge would help to discourage wasteful uses of power, and
to defray the cost of installing new generating capacity to meet demand
growth, thus avoiding step increases in base charges.) Of course, modern
computer techniques have reduced the cost of meter-reading and billing, in
proportion to other power-system costs, far below what they were in the 1950s.
Part of the reality behind the phrase is that fuel costs are a small
proportion of the total charges associated with a nuclear generating station,
essentially because the quantity of fuel required is itself very small —
tens of tonnes annually for a typical light-water reactor, as opposed to
hundreds of tonnes an hour for a coal plant of comparable output. As a
result, changes in fuel cost have only a small ultimate effect on the
cost of power from such a station. If this is true with present technique,
which typically obtains no more than 1% of the energy ultimately available
from uranium, it will be far more true with the use of breeder reactors, which
can “burn up” essentially the full fuel value of uranium and thorium both.
These two elements are not only reasonably abundant in the crust of the
Earth, but ubiquitous : there is scarcely a soil which does not contain at
least a trace of each. The main problem is finding easily-extracted
concentrations, but nature provides some help with that. In the presence of
oxygen, uranium forms water-soluble ions such as
(UO2)++, and as a result, every rain washes some
quantity down to the sea. It appears, in fact, that seawater is effectively
saturated with uranium at the present concentration of 3.2 micrograms per
kilogram, and that the total of about five billion tonnes in the
oceans of the world is static (at least over a period short with respect to
the half-life of 238U), with the annual addition balanced by the
amount carried down to the seafloor in sediments.
Thin soup, to be sure, but considerable progress has been made (especially in
Japan, where seawater is plentiful as little else is) with ion-exchange
techniques, which rely on chemical affinity to separate out the desired
species. A similar step has long been part of the process of extracting
uranium from ore, and projected prices of uranium so collected are not
completely out of sight of more conventional sources. Obviously, such an
operation could be more efficaciously performed upon brines left over from
desalination, or the petroleum brines from which we now obtain magnesium and
bromine. Equally obviously, there are other sources of nuclear fuel which can
be considered, such as many granites, some shales, and even the solid waste
from coal-burning power plants. But for our purposes, it suffices to observe
that a substantial fraction of the uranium in the ocean should be accessible
at a price which breeder plants can well afford.
According to the International Energy Agency, the world total primary energy
supply in 2011 was 5.5×1020 joules. Assuming breeder reactors
which convert the fuel value of uranium into useful energy with an overall
efficiency of only 10%, each kilogram yields up 8.1×1012 J, and to
supply three times present world energy calls for 2.0×108 kg of
uranium each year, or in the course of a thousand years, about 4% of the
quantity dissolved in the oceans. From a fuel standpoint, at least, it
appears that we may reasonably term atomic power “sustainable”.
Note : after I composed this little piece, it was pointed out to me that
much the same had been argued
by the great Bernard Cohen, among
others. But such an important matter justifies restating, and demands
repeating.
