Nuclear Fission Energy - today, in
2013 and thereafter?
In Switzerland there are four nuclear
power plants, at Beznau, where there are two reactors, at Mühleberg, Gösgen
and Leibstadt, which together provided around 3.2 GW * of generation capacity.
Nuclear generation provided 39% of the country’s electricity (26.3 TWh
out of 67 TWh), while hydroelectricity produced 55% and thermal and other plants
the remaining 5%. So nuclear provides a significant part of the Swiss
consumption.
In France in 2008 there were 58
normally working reactors which generated 76% of the country’s electricity
(418.3 TWh out of 549.1 TWh). However, at the close of 2009, a third of its
reactors are down due to ageing of components. In the summer riverside reactors
were shut down due to high river water temperatures. So the results for 2009 are
likely to be down on those for 2008. It could be said that France is
over-dependent on nuclear.
In Germany in 2008 there were 17
operational reactors which generated 29% of its electricity (140.9 TWH
out of 488.8 TWh).
In the UK there were 19 operational
reactors which generated 13.4% of its electricity (52.5 TWh out of 390.3 TWh).
And in the US there were 104
operational reactors which generated 20% of its electricity (809.0 TWh out of
4,115 TWh).
World nuclear generation amounted to 2,601 TWh
from 435 reactors providing 370 GW of capacity in 2008, which was 13% of total
electricity generation of 20,202 TWh. (1)
For 2008 the BP Statistical Review 2009 (2)
gave the world total primary energy consumption as 11294.9 million tonnes of oil
equivalent or 473 EJ, of which nuclear contributed 26 EJ or 5.5% of primary
energy. As electricity this is 9.4 EJ or just 2% of the primary energy, but this
is not a valid comparison, because the secondary energy from other sources is
subject to efficiency losses in its use.
Even if the claimed low carbon nature of
nuclear power is valid - it is dependent on many factors which show it to be
otherwise - for nuclear power to make a significant reduction in global carbon
emissions, it would mean a massive and universal increase in new reactor build.
As nuclear contributes but 13% of the world’s electricity, with hydrogeneration providing 17%, to replace the 70% generation from fossil fuels by nuclear would require 5.4 times more nuclear generation than the present capacity, let alone cater for growth in consumption.
The hypothetical natural uranium requirement
for all the current electricity generation to be nuclear (apart from hydro)
would be 352,000 t U/year, compared to the present 65,405 t U/year.
Natural uranium (equivalent) requirements
2009
The World Nuclear Association has forecast the amount of natural uranium required in 2009 (3).The world demand is 65,405 tonnes, while the principle users are the US with 18, 867 t U, France 10,569 t U, Japan 8,388 t U, Russia 3,537 t U, South Korea 3,444 t U and Germany 3,398 t U, taking up 74% of the supply.
The caveat “equivalent” is used because
half of the US requirement is as uranium hexafluoride from Russia under the
Megatons to Megawatts agreement, whereby ex-weapons highly enriched uranium (HEU)
is diluted with enrichment tails pre-enriched to 1.5% prior to both being
blended in a gaseous form to the correct fuel enriched level of 3.5% to 4.5%
U-235. (4)
Mining production in 2008 provided 43,930 t
U, 67% of the 65,405 t U requirement, with a further equivalent 2% from mixed
oxide fuel (plutonium and used fuel or MOX), leaving a further 31% from
undefined sources, some of which must come from declining stocks, some from
reprocessing. A third of the current requirement is therefore designated as
secondary sources, the mining output being the primary source.
Natural uranium (equivalent) requirements
2013
According to the World Nuclear Association (5) by 2013 43
new reactors coming into operation will provide an additional 40,000 MW of
generation. This will require an additional 8,000 t U plus 3,000 t U to make up
the initial core charges in 2013.
So in 2013 the total natural uranium production
requirement will be 65,000 + 11,000 = 76,000 t U, an increase of 76,000 –
43,000 = 33,000 t U over 2008, assuming the secondary sources are exhausted by
then and the construction and commissioning programme is held.
However, although the 43 new reactors may be commissioned, in the US some 31 reactors, one in Finland, 17 in Japan and 3 of the 5 in Switzerland will have passed their 40 years operational life by 2013 and will require upgrading or to be permanently shut down. In France many will be temporarily shut down for their 30 years’ inspections, but will presumably be licensed for their remaining 10 years.
The need for upgrading will depend on the
past replacement of major components. Around 200 reactor vessel heads and steam
generators have been subject to corrosion and replaced. For example in France,
54 of the 58 reactor vessel heads have been replaced. Replacement of steam
generators with large tube bundles has been especially difficult as they weigh
around 600 tonnes and holes have had to be cut in the sides of the reinforced
concrete containments to allow the exchange. (6) It is also likely that
inspectors will require the control systems to be revamped as the control
elements and instrumentation will be 40 years old and out-of-date.
There is therefore much uncertainty about
the primary natural uranium requirement in 2013 as it depends on the completion
of the construction programme, the level of suspension of generation during
upgrading and whether in the aftermath of the Megatons to Megawatts deal there
will be some remaining secondary sources.
The year 2013 is significant because the
amount of available secondary uranium sources is uncertain due to the
termination of the current US/Russian Megaton to Megawatts ex-weapons dilution
deal which has provided the equivalent of 10,000 tonnes of natural uranium to
support half of the US nuclear generation. There is a continuation deal by which
Russia is allowed to sell nuclear fuel to the US and some generators have signed
up for supplies. However Russia’s consumption and mining output are roughly in
balance, but it has agreed to supply uranium to India, China and others, so
market forces may deny the US and others of much of their fuel.
The US/Russian deal only accounts for half
of the secondary supplies, while a further equivalent 2000 tonnes can be
attributed to the manufacture of mixed-oxide fuel (MOX) from recovered plutonium
and uranium from spent fuel. The remaining 8,000 tonnes must have been gleaned
from operator stocks, which presumably are in decline. (7) Recovery from tails
is limited by the world capacity in enrichment separative work units (SWU), much
of which is needed to re-enrich tails to1.5% from 0.3% U-235 to be suitable to
blend with ex-weapons highly enriched uranium (HEU).
The likely outcome is that uranium fuel
supplies will not increase to the 76,000 t U/year desired level, but that a
supply/demand deficit will be averted by plant closures and by the failure of
the reactors under construction to be commissioned to the WNA schedule.
However, the so-called “renaissance”
will still be in its political womb.
Thereafter
Cost overruns
To forecast what happens in the West after 2013, depends
to a certain extent as to what happens between now in 2009 and then. Six
examples of Generation III reactors are under construction, an Areva
Evolutionary Pressure-Water Reactor (EPR) in Finland, one EPR in France and two
EPRs in China, while two Westinghouse AP1000s are under construction in China.
The first EPR in Finland is expected to be commissioned in 2012, having been
subject to delays and overexpenditure. The contract for the two EPRs in China
was allied to the purchase and opening of a uranium mine, Trekkopje in Namibia
by Areva, 35% of the output being dedicated to the Chinese projects.
Westinghouse claims that its two AP1000s in China are on schedule.
What is crucial to the financing of the new build is the
eventual technical specification of the new build reactors, which is dependent
on the assessments of the nuclear inspectors, those in Finland being STUK, in
the UK it’s HSE/NII and in France it’s ASN. The Finnish inspector’s
scrutiny is having huge cost implications in terms of component quality and
control system architecture.
There are other significant cost implications. Apart from
the reducing ore grades for uranium fuel, the components of the alloy steels
come from increasingly lean ores, the prices of which are escalating. Due to the
irradiation-enhanced stress corrosion cracking of some tubing, an alloy steel
with more chromium is now employed. There are currently few vessel fabricators
able to manufacture the reactor pressure vessels to an acceptable standard.
A perhaps ignored factor is the imposition of carbon
taxes on the emissions from the fossil fuels employed in the manufacture of the
reactor components which will add to the cost of the components. The rising cost
of the diesel used in the civil construction work is perhaps the crucial factor,
because 180,000 tonnes of the 200,000 tonnes of an EPR are concrete and
reinforcing.
In 2003 when the contract for the first EPR was signed
the capital cost was US$ 3,000/kW generated, in 2007-8 it was reckoned to be US$
5,000 kW, while in 2009 it is probably about US$ 7,000 to 10,000/kW. During the
course of the construction materials and labour costs will be subject to
escalation, the effect of which on the costs is especially significant if the
construction time is doubled as at Olkiluoto in Finland. Areva is suffering
greatly from the agreeing of a fixed price contract, so that further reactor
prices will be subject to escalation clauses, a great disincentive to generators
wishing to invest in nuclear generation.
The return on capital invested is greatly reduced if the
construction period is subject to delay, because progress payments are made by
the client at fixed achievements in a project, for which there is no return.
The ten new build starts in 2008 were all funded by state
institutions and it is evident that there will be no new build without state
subsidy or perhaps renationalisation of privatised generators.
Uranium mining
The OECD/NEA "Red Book" (8), concludes that "… sufficient nuclear fuel resources exist to meet energy demands at current and increased demand well into the future".
The requirement is to fuel an expanding nuclear generating
sector, not merely to maintain the status quo. It has never been
questioned that enough uranium exists; the question is whether, as ore
grades progressively decline, the aggregate of the energy inputs into the
subsequent nuclear fuel cycle exceed the energy in the produced electricity.
The mining industry defines a “cut-off” point whereby
at a minimum ore grade a mine development is deemed to be viable. But the ore
grade, though the main decisive parameter is not the only criterion. Not least
is the depth of the overburden and hardness of the rock to be shifted and
milled. The presence of water-bearing sandstone has been a problem at a
high-grade mine.
The ore grade is nevertheless the most significant
parameter because the bulk of the world’s uranium deposits are low grade. For
instance in Australia, which claims to hold the largest deposits, the average
grade is 0.045% U3O8, which without co-products from a
combined ore would be below the “cut-off” grade. The prospective mine with
the lowest ore grade so far to be developed is Areva’s Trekkopje mine in
Namibia where it is around 0.015%. In this case the overburden is around a metre
and it is hoped that an alkali heap leach process will allow a reasonable
extraction yield. The normal economics of the development were subverted by the
need for Areva to close a contract for two EPRs in China whereby 35% of the
output is to be ceded to China. Areva paid US$ 2.5 billion for the mine
prospects and will still need to spend another US$ 1 billion to fund its
development.
For an economic fuel cycle, ore grades well above the
“cross-over” grade which is defined as that needed to yield an acceptable
operating margin to justify the investment in mines, fuel manufacture,
generating plant, decommissioning and waste management. Paradoxically the Cigar
Lake new mine in Canada with the world’s richest ore grade is flooded with
water from saturated sandstone above and below the workings. The water was to be
held at bay with ground refrigeration, but although now plugged, the water from
two breaches has so far proved impossible to pump out. The output thought to be
7,000 t U per annum is not anticipated until 2012, but the mine may have to be
abandoned on safety grounds.
In Australia the Olympic Dam underground mine is running
down with lowering grades in its combined ores of copper, uranium, gold and
silver and is due to be replaced by an open pit expansion. The deposits are 300
metres below the surface and it is anticipated that it will take five years of
excavation of 2 billion tonnes of rock to reach the first ores, starting in
2014. As Australia is a net importer of the diesel needed to fuel the excavation
machinery, the cost of which from 2014 to 2019 is somewhat imponderable, while
the grades of the combined ore are low, it is unlikely that the expansion will
ever go ahead.
Individual mines exhibit a build-up, a plateau followed by
a decline in production. To maintain a regular supply of uranium there needs to
be the opening of a succession of new mines. The failure to open Cigar Lake in
Canada and the Olympic dam expansion in Australia has led to the rapid fall-off
in production. In 2008 Kazakhstan’s production (8.521 t U) overtook
Australia’s (8430 t U) and may well overtake Canada’s (9,000 t U) in 2009.
Kazakhstan’s rise may be short-lived as it is based on
in-situ leaching (ISL) of a number of small mines. ISL allows a rapid build-up
of production, but then leads to a rapid depletion. The situation in Kazakhstan
is confused, with the previous management of the state enterprise imprisoned.
Since 2005 Canada’s uranium production has fallen 22.6%, while that in
Australia fell 11.4%.
The US and French mines’ production curves are classic
Hubbert. US uranium mining provides only 10% of the country’s demand and the
French mines are exhausted.
What the “Red Book” describes as “World uranium
production capability” should now by now be exploited to produce 80,000 to
87,000 t U in 2010, whereas it is likely to be only half of it around 45,000 t
U.
The nuclear proponents often claim that the amount of fuel
needed for nuclear generation is miniscule. A 1 GW coal-fired station consumes
around 2.87 million tonnes of coal a year, whereas a similar nuclear output is
obtained with around 200 tonnes of natural uranium. At 0.045% grade this amounts
to 200/0.848/0.00045 = around 0.52 million tonnes of ore, which with a waste
rock/ore ratio of 5:1 (as in a deep open pit) means 2.6 million tonnes of
material has to be shifted and processed; a somewhat similar amount.
Fast breeder technology
There are arguments as to how much uranium
can be anticipated from mining, but little doubts as to its ultimate economic
exhaustion. So apart from perhaps a hope that it can be extracted from
phosphates and seawater, in the OECD/NEA “Red Book” (8) is the claim “Moving
to advanced technology reactors and recycling fuel could increase the long-term
availability of nuclear energy from hundreds to thousands of years”.
But just how the transition from
“once-through” fuel technology to a renewable breeder cycle is not defined.
The “Red Book” argues that “There
is enough uranium known to exist to fuel the world's fleet of nuclear reactors
at current consumption rates for at least a century” so that if that is
the case there is no need to consider the introduction of fast breeders for some
time. Indeed the agency has admitted that there has been no attempt to model the
transition.
However if there is to be a
“renaissance” and nuclear generation is to be doubled or tripled, the
century is reduced to 50 or 30 years, which is incompatible with the 60 years
declared life of the generation III reactors.
Fast reactors can be set up as
“burners” to destroy actinides including plutonium or as “sustainable”
with no gain or loss of plutonium or as “breeders” with perhaps a 12% gain
per year. So at best the initial plutonium charge would be recovered in 10
years. The breeder is one part of a three-stage process of the separation of
the recovered plutonium, the manufacture of the plutonium and depleted
uranium fuel and the actual breeders, which would presumably be at least ten in
number, fuelled and discharged in sequence. Its understandable that such
modelling is deferred as the sequence will probably never work as it is a sort
of “perpetual motion”.
To substantiate its claims of
long-term availability it should be demanded of OECD/NEA that it supports its
claims with a mathematical model.
Dry casks
Spent fuel is accumulating in ponds
associated with pressure and boiling water reactors, so that once the
radioactivity has decreased (after 10 to 20 years) it can be transferred to dry
casks. In the US the casks are stored in outdoor “cemeteries”, while in
Switzerland and Belgium they are stored in “mausoleums”. With the suspension
of building of the Yucca repository, it appears that in the US
the dry casks will remain in guarded storage areas well into the next
century. This appears to be the solution in the UK, as the Sizewell B spent fuel
pond will be full by 2015, so to allow generation to continue for a further 20
years thereafter, the process of removing the oldest spent fuel and placing in
dry casks is to begin, subject to planning approval.
The transfer may be overtaken by events, as
several spent fuel ponds are leaking. Also the cooling and filtering of the pond
water by circulation has to continue for 10 to 20 years after the generation has
ceased, so that a standby source of electricity is required from 2080 to 2100 if
the 60 years operational life claimed can be sustained. How this essential
energy supply will be guaranteed at the end of the century is uncertain, but if
it is not and the ponds dry out and the fuel catches fire, the consequences can
only be imagined.
The alternative of reprocessing may be
effective in France and Japan, but is absent in the US, while the Thorp plant in
the UK has been a disastrous failure and is to be decommissioned.
Low carbon?
The nuclear “renaissance” has been
promoted as a source of low carbon energy and its life cycle specific emissions
are claimed to be as low as 5 g CO2/kWh, comparable with wind power.
However, the figure is based on the aggregate emissions over its life cycle,
with three main emission components, viz., the emissions during the manufacture
and construction, the emissions associated with the fuel mining and manufacture
and the emissions related to the spent fuel management.
The problem for this analysis is that a
succession of building will release the greatest amount of the emissions for
several decades. Thereafter the recourse to lowering grades will increase the
emissions related to fuel manufacture. So even if the life cycle emissions are
low, the benefit for climate relief will not be realised until it is too late.
Also the low figure of 5 g CO2/kWh has been disputed and may lie
between 60 and 150 g CO2/kWh, increasing to 300 g CO2/kWh.
(9)
Also if the operational life is curtailed for any reason,
such as a lack of cooling in the summer, the failure of a major component, or
even a lack of fuel, the specific emissions rise accordingly.
Due to the falling uranium mining production
in Canada and Australia together with the ending of the US/Russian ex-weapons
deal, nuclear generation in the US and France will decline. Areva will fail to
supply its hegemony with fuel and will have difficulty in providing the initial
core charges of its new build, which are 2.3 times bigger than the subsequent
charges. The ephemeral rise in production in Kazakhstan will be to the benefit
of those currently holding supply contracts, viz., China, Russia, South Korea
and Japan.
The catastrophic rise in costs and the construction delays at Olkiluoto will prevent the financing of the “renaissance” by the private sector. The current programmes of new build in Asia and particularly China, though state funded, will suffer escalations in costs and will be delayed. Some new build in the West may be subsidised.
Depending on how many new stations are actually commissioned and when, the nuclear fuel supply may eke out, but as uneconomic low grade mines have to be opened to “keep the nuclear lights on”, the price of nuclear electricity will rise as will other energy sources.
Nuclear power will, with economic activity, fade in
accordance with the depletion of fossil fuels, without which to mine its uranium
and fuel its construction and maintenance, its “renaissance” cannot succeed.
John Busby 9 December 2009
As a supplement the hydrogen economy is
examined below.
Hydrogen-Fuelled Transport
The world’s mobility currently depends on petroleum products, resorting to gas and coal to liquids processing as oil production reduces. In the remote windswept Shetland islands wind generators are providing hydrogen for small vehicles. Storing liquid hydrogen takes care of the variability of wind. In Reykjavik buses are fuelled with hydrogen produced from geothermal energy. Otherwise hydrogen is mostly produced from natural gas, discarding its carbon. Could hydrogen from nuclear electricity by electrolysis keep the world moving?
Assuming mobile transport requires 40% of global primary energy (473 x 0.4 = 189 EJ) and taking into account the energy efficiency of a diesel engine at 46%, the useful energy is 87 EJ. Taking the lower heating value of hydrogen (120 MJ/kg), the mobility energy is carried by 725 x 109 kg H2, which to get in its liquid form requires by electrolysis and liquefaction 75 kWh/kg equivalent to 54,375 TWh, which at 34% generation efficiency requires 576 EJ of primary energy.
So if transport could be based on a hydrogen fuel the primary energy requirement for it rises from 189 EJ to 576 EJ, a rise of a factor of 3. A 1 GW nuclear power plant provides 85 PJ per annum, so if petroleum based transport fuels were to be replaced with hydrogen produced by nuclear power it would require the building of 7,0001 GW stations (or 4,400 1.6 GW EPRs) requiring around 1.4 million tonnes of natural uranium per year. With the need to consume 3 times the primary energy from petroleum-based fuels the universal application of nuclear generated hydrogen fuelled transport is unlikely. Also the nuclear power plants would have to be located near the centres of transport activity.
One of the development candidates in the Gen IV Roadmap (10) is a gas-cooled high temperature reactor with an ability to produce hydrogen, but its successful implementation, given the lack of materials able to withstand the high temperature and the need for an hydrogen infrastructure is unlikely.
Hydrogen fuelled transport would require a huge distribution infrastructure together with alternatively powered vehicles. It is entirely inappropriate for developing countries, where only diesel powered trucks can deliver goods. The electrical alternative to hydrogen by electrolysis is to employ batteries, which may well be applicable to small passenger vehicles, but is useless for bulk transport due to the weight penalty.
The only conceivable solution (to substitute for petroleum for road transport with electricity) is to lay rail tracks with overhead line collectors on emptying motorways. It is possible to fuel aircraft jet engines with hydrogen, but would need the provision of an infrastructure at all participating airports as otherwise arriving hydrogen-fuelled flights would be grounded. Moreover, the consequent redundancy of existing kerosene- fuelled fleets by an industry in decline (following the passing of a peak in oil production) could not be managed.
Universal hydrogen-fuelled transport is an unlikely prospect.
Reference
documents
(1)
IAEA PRIS database http://www.iaea.or.at/programmes/a2/
(2)
BP Statistical Review 2009 http://www.bp.com
(3)
WNA Reactor table http://www.world-nuclear/info/reactors.html
(4)
WISE Uranium Downblending
Calculator http://www.wise-uranium.org/nfcub.html
(5)
Plans for new reactors worldwide http://www.world-nuclear.org/info/inf17.html
(6)
Major component replacement http://www-pub.iaea.org/MTCD/publications/PDF/Pub1337_web.pdf
(7)
Uranium Markets http://www.world-nuclear.org/info/inf22.html
(8)
OECD/NEA “Red Book” http://www.nea.fr/html/general/press/2008/2008-02.html
(9)
Storm van Leeuwen http://www.stormsmith.nl
(10)
Gen IV roadmap http://gif.inel.gov/roadmap/
* The units used in this analysis will be
confined to joules and watts, expressed in exajoules (EJ), petajoules (PJ),
megajoules and terawatts, gigawatts,
megawatts and kilowatts.
Exa = 10^18
Peta = 10^15
Tera = 10^12
Giga = 10^9
Mega = 10^6
Kilo = 10^3
In the US, a unit known as the quad
(quadrillion British Thermal Units) is roughly equivalent to an EJ).