The problem of estimating the ERoEI of nuclear energy has
arisen once again. In 2005, I read the
MIT report and the University
of Chicago report and, as
part of a much larger project, wrote as follows:
Regardless
of the finiteness of uranium resources, nuclear energy must be considered
renewable because of the existence of fast breeder reactors and the likelihood
that their technological limitations will disappear over the coming
decades. Therefore, nuclear power should be admitted to the competition
with wind, solar, biomass, and other sustainable technologies. If there
is some reason why nuclear energy is not sustainable, it has yet to be
demonstrated. (What is not sustainable is growth itself – not nuclear
energy.)
Suppose
that we agree that the hydrogen economy means hydrogen from nuclear power
installations (NPIs). Suppose that we agree that the hydrogen economy
means hydrogen from nuclear power installations (NPIs). [However, see [
http://www.phoenixprojectpac.us/user/Phoenix%20Project%20for%20America%20PAC.pdf]
for a non-nuclear approach to the hydrogen economy.] In their article
“Large-Scale Production of Hydrogen by Nuclear Energy for the Hydrogen Economy”
[
http://web.gat.com/pubs-ext/MISCONF03/A24265.pdf],
K.R. Schultz, L.C. Brown, G.E. Besenbruch, and C.J. Hamilton suggest that
hydrogen can be produced with a 50% efficiency by thermal splitting of water
using a Sulfur-Iodine cycle in conjunction with the Modular Helium Reactor
(H2-MHR). The efficiency of the H2-MHR bypasses the objections to using
electricity as an intermediate step as discussed by Ulf Bossel, Baldur Eliasson
and Gordon Taylor [
http://www.oilcrash.com/articles/h2_eco.htm].
Other drawbacks of hydrogen have been addressed by Graham Cowan in his
interesting paper Boron: A Better Energy Carrier than Hydrogen? [
http://www.eagle.ca/~gcowan/Paper_for_11th_CHC.html]
Also,
associated with the hydrogen economy and whatever residual industrial tasks
cannot be converted to electrical power are the huge changes in our
technological and industrial infrastructure associated with conversion to the
use of hydrogen for fuel. This will involve energetically costly
re-tooling for the production of different types of industrial equipment.
Although the period of amortization can be prolonged, ultimately such costs
must be charged to the energy invested in nuclear energy.
The
cost of liquefying hydrogen might be paid in part at least by using hydrogen to
facilitate transmission of electricity through ‘high-temperature’
superconducting transmission lines that might run through the middle of liquid
hydrogen pipelines. I do not know if this is feasible nor do I have a
reference for it as I have no idea if it exists outside of my own
imagination. However, I have noticed that the fractional losses of
electric power listed in the reference case from
Annual Energy Outlook 2005 (Early
Release) (AEO2005), published
by the Energy Information Administration of the US Department of Energy (DOE),
are rather large so that the potential savings, at least, are documented.
(See Appendix A of
AEO2005Full.pdf.)
[Note. The term ‘high-temperature’ means that, while the temperature is
still cryogenic, it is well above absolute zero.]
If
the Energy Returned by NPIs is less than the Energy Invested, nuclear energy is
infeasible. Therefore, the frequently discussed ER/EI analysis is crucial
to this discussion. Probably, the ER/EI ratio for nuclear power is less
than comparable ratios for fossil fuels, which is a drawback insofar as market
penetration is concerned; however, so long as it exceeds 1.0 the introduction
of nuclear energy is feasible. There are a number of factors, however,
that point to the possibility that ERoEI is less than 1.0. In particular,
elsewhere in this section, a number of requirements of NPIs are mentioned
that might be easy to overlook in an analysis of ER/EI.
The
identification and quantification of every component, both direct and indirect,
of the energy invested in nuclear power is not a simple thing to do. In
particular, if any such study of Energy Invested includes the ancillary
business expenses, including the expense of doing the very study in question, I
have not seen it. But, in the American economy, for example, the energy
consumed by commerce is 22% of the total energy budget. This is corroborated
by employment statistics. (See [
http://stats.bls.gov/oes/home.htm].)
However,
it is not clear that all ancillary costs have been included, e.g., desalination
of sea water, remediation of environmental change, etc. A pro-rata share
of the costs of providing and maintaining railways to carry heavy equipment,
fuel, and waste, highways to transport workers, conduits to transmit electric
current, pipelines to transport hydrogen, and easements through which
electrical power lines and hydrogen pipelines can be run must be charged to the
plant. Some locations for NPIs are unsuitable for this necessary
infrastructure, and, therefore, unsuitable for NPIs.
At
the start of this exercise, I considered the notion that I might be able to
determine the feasibility of nuclear by looking at the energy balance for
France.
(
http://www.eia.doe.gov/emeu/world/country/cntry_FR.html)
France produces about three
quarters of her electricity from nuclear, but
France has to import about half of
its energy. Is it possible that nuclear power consumes more energy than
it produces? Despite the inclination to prove the affirmative, I have not
been able to determine the answer to this question by looking at the available
data. In fact,
France
seems to be doing rather well insofar as energy is concerned; and, therefore,
is much less of a problem for the rest of the world than is the
United States.
Finally,
and we shall have to await a more thorough discussion of this topic, the author
wonders if the cost of restoring the land and the water employed by NPIs to its
pre-nuclear condition should be charged to the Energy Invested even if
there is no possibility that the land will ever be used for any other purpose
than nuclear power into the foreseeable future. Clearly, decommissioning
costs must be included, but does decommissioning include restoring the land to
its original condition as a beautiful, natural, wildlife habitat? Quite
frankly, I believe that it does.
Although
the capital costs of NPIs are sufficiently high that market penetration under
the standard short-sighted micro-economic model might be prohibitively
difficult, as a fraction of the projected Gross Domestic Product they are quite
manageable by a society that possesses the political will to manage them as we
shall see in the sequel. The final irony might be that a capitalist-style
market economy can be maintained under a centrally-planned socialist energy
economy and only under such an economy.
Many
people believe that the United
States economy is in such bad shape,
principally because of the trade deficit and the national debt, that it could
not possibly support the massive spending necessary to install a hydrogen
economy. If the government continues to run a deficit, the public costs
of such a project might very well multiply that deficit by a large
factor. While this may be true, it does not necessarily represent the
prohibition of the Apollo Plan, so long as American workers are willing to
accept government debt in the form of fiat money as payment of wages.
This study shows that capital costs are well within the capabilities of the US
economy. The results are presented as the final two computations done on
the spreadsheet explained in the body of this report.
Unfortunately,
nuclear facilities are operated sometimes for the personal profit of their
owners, managers, and other stakeholders who might be inclined to place their
personal interests ahead of other considerations such as good engineering
practice and safety.
Mere prudence dictates that we be suspicious of enterprises run for
profit. Since it will require huge investments by the federal government
to penetrate a market economy with current nuclear technology, the federal
government might just as well own and operate whichever nuclear plants it
chooses to subsidize. The Apollo Plan amounts to some sort of Socialism;
hopefully, it will not be Corporate Socialism, i.e., Fascism. Thus, the
evils of the profit motive can be avoided, but only by compromising
Capitalism. However, as critics of Socialism will be quick to attest,
this does not necessarily protect society from incompetence.
Nuclear
Power Installations (NPIs) need fresh water. Many experts believe that we
are even closer to Peak Water than we are to Peak Oil if we are not past
both. Since some experts disagree, this must be regarded as an open
question. If fresh water is used as cooling water, it must be returned to
the environment at the original temperature with all contaminants removed and
all nutrients restored. If fresh water is split to produce hydrogen, it
may end up as atmospheric water only part of which will return to Earth as
fresh water, in which case the losses in our fresh water supply will have to be
replaced somehow. If some of our NPIs are used to desalinate sea water,
the energy expended must be subtracted from the Energy Returned in computing
ERoEI.
As
an example of water use by an existing nuclear power facility, nuclear Plant
Hatch in
Georgia withdraws
an average of 57 million gallons per day from the
Altamaha River
and actually "consumes" 33 million gallons per day, lost primarily as
water vapor, according to the U.S. Nuclear Regulatory Commission (
http://www.cleanenergy.org/programs/water.cfm).
Plant Hatch, consisting of two 924 MWe reactors each with a capacity factor of
0.8453, consumes water at the rate of 3.2903 x 10
11 kgs/emquad. Thus, if every NPI
in the year 2100 used water at the rate Hatch Plant did in 2000, we would need
1.1442 x 10
15 kgs of
water per year to satisfy the modest economic growth assumed in my
Reference
Case for the
Conservation-within-Capitalism Scenario. According to
http://www.american.edu/TED/water.htm,
we have about 3 x 10
15 kilograms
of renewable fresh water total. Thus, power plants would use more than
one-third of all of our renewable fresh water. According to
http://oldfraser.lexi.net/publications/critical_issues/1999/env_indic/resource_use.html,
the
US
has 2.5 trillion cubic meters of water or 2.5 x 10
15 kgs, which corroborates the previous
estimate. Also, see
http://www.worldwater.org/table1.html.
Some
of the energy produced can be used to desalinate sea water for reactors on our
East, West, and South coasts where the population is dense and fresh water
dear. Moreover, energy from ocean waves can be used to assist
desalination. [
http://www.malibuwater.com/OceanWaveEnergy.html]
Let us compute a lower bound for the energy cost of desalination of sea water
to make the case against nuclear as conservative as possible. According
to Allan R. Hoffman (
GlobalWater.htm), “energy requirements, exclusive
of energy required for pre-treatment, brine disposal, and water transport, are:
reverse osmosis: 4.7 – 5.7 kWh/m
3and multi-stage flash: 23 – 27
kWh/m
3”. To establish a minimum, I shall use 4.7 kWh/m
3 to obtain
i.e., an increase in Energy Invested of 1.6% of the Energy
Returned, which should not present a problem. However, if the higher
value for multi-stage flash were the best one could do, the costs would soar to
nearly 9% of the Energy Returned. If the ratio of Energy Returned to
Energy Invested (ERoEI) were 5.0, the energy costs would increase by 44.9% and
the ERoEI would be reduced to 3.45, which would certainly be an unwelcome surcharge
on nuclear power. In addition to the costs of pre-treatment, brine
disposal, and transport, the cost of desalinating water to be split into
hydrogen and oxygen would have to be borne. The cost of transport might
be considerable if sea water were needed in Minneapolis, say, but the scarcity of fresh
water is most acute in places much closer to the ocean. The calculation
of these additional costs shall be postponed to some future study.
The
final limitation upon economic growth is the area of the surface of
Earth. NPIs require a smaller fraction of Earth’s surface per unit of
power generated than any of the competing technologies, namely, wind, solar,
and biomass – despite the fact that solar and wind power installations can
coexist with other land uses. Even if every other obstacle to growth were
removed, ultimately we should run out of space – unless some means of
miniaturizing NPIs, for example, should be discovered such that the rate of
increase of power density could keep pace with growth. (If emquads per
square meter increases at the same rate as emquads, we would be able to produce
the energy budget of the future in the space we use now.) Even in the
unlikely event that NPIs could be stacked, a limit would be reached after which
they could be stacked no further without the expenditure of more energy than an
NPI can produce during its lifetime. Also, there are limits to power
density that, if none other could be found, would be set by the atomic nature
of matter – although, admittedly, if the concentration of the space per unit of
power were limited by atomic considerations alone, growth might continue for a
very long time. Probably, though, by the time the individual Earthling
could wear an NPI strapped to his wrist like Dick Tracy wore a radio, we shall
no longer be living on Earth, a situation to be deplored for other reasons as
stated previously.
To
return, for a moment, to more realistic considerations, the land needed for
NPIs includes not just the plant sites and infrastructure for transportation
and power transmission but also the space occupied by facilities for mining and
enrichment, fabrication, maintenance, recycle, hydrogen compression and
liquefaction, waste management, sea water desalination, fresh water
remediation, and the ubiquitous office buildings that seem to be a necessary
part of every enterprise engaged in the pursuit of profit. Engineers and
scientists will need workplaces; and, if I am not mistaken, the greater the
complexity of our energy economy the greater the superstructure of command and
control, which, in the case of nuclear, must be multiply redundant.
Moreover, many areas on the face of the Earth are not suitable for NPIs,
namely, the tops of mountains, earthquake zones, crowded cities (perhaps), and,
if we wish to observe the ethical treatment of animals, wildernesses, swamps,
prairies, etc. – in short, any
place where humans have not yet evicted animals from their natural habitats,
which, for all practical purposes, amounts to saying that future nuclear
installations may be placed nowhere. Finally, it must be decided
whether the space occupied by outmoded and obsolete facilities can be reused
for new facilities or if it must be restored to the pristine condition in which
Nature bequeathed it to us. If the latter, the energetic costs will very
likely overwhelm the Energy Returned in the ratio (or difference) represented
by ERoEI, which brings me to the next point:
Quite
obviously, while operating as designed, nuclear power plants do not contribute
directly to Global Climate Change nor air and water pollution regardless of the
effect of their ancillary facilities, e.g., mining, etc. When nuclear
facilities are operated properly, the dangers are rather minimal; nevertheless,
nuclear radiation is extremely dangerous. In addition to radiation
poisoning, nuclear plants have a non-zero, but very small, probability of
exploding; but, if there are many of them, the probability of explosion
increases accordingly. Admittedly, there is no physical reason why the
problems associated with pollution, radiation, explosions, waste, and
decommissioning cannot be solved, however they must be solved; and, to the
extent that they have not yet been solved, they represent impediments to the
introduction of nuclear power and the hydrogen economy, which brings us to the
next topic.
Nuclear
power is the key to a much larger and complicated economy with much greater
opportunities for unanticipated environmental catastrophes both because it
makes a larger economy possible and because it makes a more complicated
economy necessary to supply an energy budget that is growing
exponentially. Now, the economy is sufficiently complicated in 2005 that
the average person must necessarily depend upon the opinions of experts to
determine which public policies are in his best interests and which are
not. Moreover, experts disagree. The average man or woman is held
hostage to the complexity of the economy, and this situation is not conducive
to democracy. Soon enough, under a scenario of modest growth, this
situation will be exacerbated many times over. The interests of ordinary
private individuals will be taken out of their own hands almost
completely. Presumably, a technocracy is better than a plutocracy (unless
technocrats become plutocrats); but, in either case, it represents social
degeneration – not progress.
[snip]
As one can see, this was written before - but just before -
I began to consider ERoEI by drawing my control surface around an entire
community, a community living in an Autonomous
Alternative
Energy
District
as in http://dematerialism.net/remarks.htm.
However, without repeating the earlier computation, one can take new
data such as (1) the Energy Information Agency's cost per kilowatt-hour
(electrical) of electricity from nuclear energy of $0.108 in 2013 from
http://en.wikipedia.org/wiki/Economics_of_nuclear_power_plants;
(2) the Gross Domestic Product of the United States in 2012, which is close
enough to 2013 for our purposes, of 16.25 Trillion US Dollars; and (3) the
Total Energy Budget of the US, which was 95.14 quads. Thus, since most components of the energy
investment contribute a corresponding component to the monetary cost, we can
compute an upper bound for the ERoEI of
0.54. Wow! I must have made a mistake. I haven't got time for Microsoft's equation
editor; so, I had better scan a pencil calculation and include it as a
jpeg. I will place this on-line in case
Google doesn't print equations or figures.
See
http://eroei.net/nuclear.htm
. The actual ERoEI* must be less than
this because society does not require some of the measures that would have to
be taken to achieve genuine sustainability; however, the technology is
infeasible as it is.
Alexander Carpenter is rather skeptical about this
computation because the levelized cost represents the amount that nuclear
energy would have to sold for to just break even. I believe that the break-even price includes
all obligatory payments to investors and lenders. We don't care if the industry makes a profit
or not; but, if it does make a profit, the additional energy consumption of
those who earn it must be charged to the technology. This is accounted for by the price. The energy associated with every single
component of the price including the taxes collected must be reflected
appropriately in the energy investment term.
This, then, is a slightly edited version of my reply to him
on the America
2.0 forum:
Rather than re-send the post I sent earlier tonight from the
wrong email account, I would like to ask Alexander to take a moment to read
this carefully. Obviously, he has a great deal to contribute to this
discussion; but, I am not interested in ROI. I may not understand what
GDP means nowadays; so, I will tell you what I think it means with the
hope of being corrected if I am behind the times: GDP is supposed to be
the sum total of all goods and services purveyed and purchased
domestically during a given year. It is the flow of money through the
economy as in Odum's famous diagram, which was first rendered when the US GDP
was only 1.4 trillion USD/year.
Next, I need to be sure what the levelized cost is and how
it is intended to be used. I think that's where Alexander got the
idea that I am after ROI. I think I should use the average price to the
consumer regardless of sector because this is how much money enters the economy
when one kilowatt-hour of electricty is sold regardless of origin. This
was 0.1045 USD according to
http://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_5_6_a which
is assumed to be a reliable source of raw historical data. In my
experience, these bureaus are staffed by serious, competent people who would
rather not be pushed around by unscrupulous businessmen. I don't think these
people are much influenced by political pressure, which is exerted much farther
downstream in the dissemination of data. Are they not tenured? I
have had serious discussions with the people who supply my data and they have
had a lot to teach me. In addition, they seem to be pleased to show the
general public how useful they are. It doesn't matter to us if the average
price to the consumer is affected by subsidies; however, since it does not
reflect things like environmental reparations that should be done but are not
done, the figure I come up with delineates feasibility rather than
sustainability as I defined these terms at
http://eroei.blogspot.com/ in
the post of August 14th, 2013. Finally, I believe we are in agreement on
the total energy budget, E. The ratio of E to GDP is the quantity of
energy flow through the economy counter-current to one USD of monetary flow.
The new value of ERoEI is higher, then, by the ratio of levelized cost to the
average price to the consumer or 0.108/0.1045 X 5.4 = 5.6. This
result is subject to the determination that 1 KWH(e) is really 1 KWH and not a
third of a kilowatt-hour, the electric energy generated by the expenditure of 1
KWH of fuel at the conventional power plant. I would like some input on
that question.
To make absolutely clear what I mean by this technique, I
illustrated it for the Mark II economy in
http://dematerialism.net/Mark-II-Economy.html
where it is crystal clear how each term was computed and, therefore, what it
means.
Tom Wayburn, Houston,
Texas