Is there enough Uranium ore to do the job?

In a recent post, Nuclear Decisions-4, we examined whether nuclear power might be able to replace all other forms of power generation.  Our answer?   No, we will run out of fuel immediately we were to try.

That answer assumes we continue current US  once-through policy – Sequester the fuel after it has been in a reactor and has generated enough reaction poisons to halt fission activity.  The grand idea is that we will put it somewhere inaccessible by humans for the next million years.

Let’s reconsider nuclear applications with a fresh viewpoint.   If we cannot replace ALL power generation, is there any fuel we could replace with uranium?

Uranium stores more energy-per-kilogram than in any other fuel

Table 1 is our chart on the energy richness of different fuels.  Footnote [1] discusses assumptions behind this table.

Fig 1 Uranium holds by far the most energy

Uranium is clearly much richer in energy than oil or any other fuel source. So we naturally ask: The other fuels are running out, or have pollution issues … could we substitute uranium for these and alleviate the problem?

Probable lifetime of uranium reserve, TU = 230 years.


Fig 2: Est lifetime of world NatU reserves

In the Annex (down below),  uranium industry data shows that the world currently uses about 71,000 t of natural uranium each year, and the EIA  [2]  says we generate 2.6 T kWh (trillion kiloWatt hours, a.k.a tera kiloWatt hours) annually.  Based on the RAR (Reasonably Assured Resource) of 5.4 Mt (mega tons) of natural uranium, we expect natural U to last 76 years if used at its current rate (constant, no change over the lifetime).

Most people in the industry would multiply the RAR by 2 or 3.  The current low prospecting level for  uranium will go up when positive returns on investment happen;  people expect large increases in potential mining sites when proper geological surveys are conducted.

If the uranium demand is to triple the amount expected ore,  the probable lifetime of mineable resources will be nearly 230 yrs.

We will use the basic 76 year value for our Expected values.  We stretch each answer by 3 for the Probable value.

Replace petroleum energy by nuclear?

Society has either reached the end of exponential increases in petroleum products, or is very close to that peak point.  Ever-increasing demand will NOT be met in the future.  This paints a grim vision of ever-increasing prices for the same (or decreasing amounts) of oil.  Ready cameras  Scene-I backstory:  crashing economies, raging wars, diseased and poverty-stricken masses,  small enclaves of ultra rich trying to stay comfortable.


Fig 3: NatU lifetime if it replaced oil

Petroleum data:   World consumption 2010 [3]:  84.5 M bl/d,  30.7 G bl/y. Energy generated: 178 Quads (quadrillion BTU aka  1015 BTU), or 52.2 T kWh  per year.

Nuclear option:    If nuclear were to take over the petroleum supply,  it must annually supply the petroleum energy, in addition to the  amount it currently  generates.  For a reserve of a fixed size,  Usage × Lifetime is constant.  Use more, the reserve lasts less,

54.8 * T = 2.6*76 …  Right side: Fig 1, Left side, new total energy × lifetime petroleum replacement, Right side is current nuclear usage

Petroleum Answer:  The probable lifetime is 11 yr.  No, nuclear resources are not an alternative for petroleum.  One decade to resource exhaustion would not be sufficient gain for the price we must pay to … build reactors, operate them safely, decommission  them after natural U is gone, and then finding  guaranteed safe storage for the waste.

Replace coal energy by nuclear?

Coal is very destructive to the environment.  Cities that use it have terrible smog.  Currently burning releases 6 billion tons(6 Gt) into the atmosphere and is a strong driver of the greenhouse effects. Temperatures are indeed rising, releasing more free energy into the atmosphere; causing more chaotic swings in environmental turbulence.  We really could use an alternative to burning our several hundred year coal reserve, especially high sulfur coal.  Could we do this with the nuclear option?

Fig 4: NatU lifetime if it replaces coal

Fig 4: NatU lifetime if it replaced coal

Coal data:   World consumption [4]: In 2009, world use was 7  M t of coal,  143 Quads of energy, corresponding to 42 T kWh in energy.  This is a very rough estimate because of huge variation in thermal energy in the various kinds of coal.

Following oil replacement ideas, 44.6 * t = 2.6*76. Expected NatU lifetime = 4.4 yrs.

Coal Answer:  No, Nuclear power is not an option to replace coal use.  One and a half   decades to resource exhaustion would not be sufficient gain for the price we must pay.  If every country decided to do this, the reserve would drain and resource wars would be probable..

Final comments

Analysis of our four previous Decision posts, indicated at least these 3 points:

  • The power generating reactors around the world are a real and present threat.
  • Fuel melt down incidents need not be inevitable.  We never need watch another one happen.  We could have built intrinsically safe reactors decades ago.
  • Sequestration strategies can not be trusted.  The time frame for safe-keep is well beyond our ability to predict.  Failed sequestration will have fatal consequences.

Our point here is that, as energy-rich as uranium may be, it can not make up for the failing energy sources that society depends on.  We are blocked from success by our once-through rule.  The root cause behind this rule is not a technical issue, but a social one: Governments do not believe they can control the overseers who gain financially from nuclear power.  This is a sad truth; because of it, we did not adopt reactors that are intrinsic safety and burn “waste” fuel.

The once-through rule plus operating reactors form the worse case scenario.  Reactor disasters will continue to occur throughout this century.  As our petroleum base becomes ever more expensive, each new incident will be harder to contain that its predecessor.  If we do not act (and soon), the present system will crumble in its weakest spots, and society at year 2100 will be significantly degraded.

All our estimates of limited uranium resources become moot if we can burn the “waste” fuel for energy; they become meaningless if we learn to safely breed fuel.  With a rich source of power to build on, we can develop maintainable lifestyles  – decent food, good medical treatment, well-paying jobs that allow us to live with pride.

This requires we build reprocessing plants to supply the new type of intrinsically safe gas cooled reactors. These new plants must be government-run and military guarded. This can work, we did this after WW-II with the National Laboratories that handles nuclear weapons information.

I admit to a bit of discomfort with this statement, probably due to personal decisions made long ago.  I decided to study plasma physics and fusion technology rather than nuclear physics and reactor technology partly from my personal distrust of the fission power politics  …but… The bedrock of hard fact remains: we cannot sustain ourselves as a society without a rich energy resource to draw on.  Fusion was abandoned, fission remains the only viable source that can support energy intense activities such as manufacturing and mining.

And we can do the job. We can extend existing nuclear options until sea water is an economic reality.  There is yet a path to a secure, sustainable and comfortable future based on the astonishing technical achievements we have made collectively during the last several thousand years.

click for LastTechAge on fission technology

We could, but we must make the decision soon. Indecision is a decision for the downward path.  Decisions to extend the status quo, even by massive amounts, would drain irreplaceable resources into inadequate (current) solutions; it would only enrich vested interests.  This this would  enhance our decline, not our future. Do we in the US have the will?  Do we in the US, Canada and Europe have the will?  Do we in…
Charles J. Armentrout, Ann Arbor
May 29, 2011, Update 2013 Apr 10, added tables
Listed under    Technology     …   Technology > Fission
Have a comment? Click on the title of this post, go to bottom, let us know.

Related posts: Click the INDEX button under the Banner picture


Annex:  Available Uranium

Previous estimate.   In Nuclear Decisions-4, we followed the lead of an  estimate for uranium that was published in the  popular science magazine, Scientific American [5].  Three parameters were needed:


Fig 1 – Uranium enrichment classes

  1. World energy use [6]
    (2.8 T kWh/yr),
  2. Power generated by LEU
    (1 t LEU produces 400 M kWh)
  3. Amount of natU needed to make LEU (10 kg natU per kg LEU).

These were sourced back to the US Department of Energy’s Energy Information Agency.  Note: T means million×million, pronounced as  “tera”  (international standards) or “trillion” (US convention).  Results – the world uses about 70,000 t of uranium each year and if we multiply the known reserves by 3 to account for undiscovered resources,  we calculate a 260 year supply in the uranium reservoir.

We will not distinguish between metric and English tons because they differ by only  10% and our estimates will certainly be no better than 10-20% of the exact value.


Fig 2–  Top 10 Uranium mining countries

Second estimate: Same question from economic viewpoint.  Uranium occurs in nearly everything [7],  with 2.8 ppM average concentration in the soil and 3 ppBillion in sea water (about 500 M tons of U).  This is so dilute that the huge majority of it is not obtainable in any cost effective way.  This differs from petroleum which exists in well defined pools.

Mineable deposits of U do exist and Fig 2 shows that the top three reserves are in Australia(31%), Kazakhstan (12%)  and Canada (9%) [8].  The US  has about 4% of the known deposits. Kazakhstan is the leading ore producer. The result of extracting natU from ore is a yellow powder compound, called yellowcake.

Fig 3 –Ore classifications

A few mines have U at concentrations of about 10% (1 kg U per 10 kg ore) or more, but most are much less.  Fig 3 shows ore classifications, which is done by concentration, which extend between 0.01% (1 kg natU per 10 tons of ore) and 10%.  On this kind of log graph, the width of the decades are the same, but steps shrink as they move up from 1 to 9.  The average is about 0.15% (1.5 kg natU per ton of ore). The kind of matrix also matters, from soft sandstone (cheap to extract) to hard granite (expensive to extract).

If concentrations are below 0.01:  (A) the ore is  uneconomical to mine, and, (B) it would be better to use the extraction fuel carbon-based to generate energy because it would produce less CO2  per kWh than would be emitted by processing the uranium needed to generate the same amount of kWh [9].


Fig 4– Demand has exceeded supply since Chernobyl

All of this is relevant to estimates of how much natU exists in usable concentrations.  Although there is a lot at low concentrations,  for power generation, we will extract only that which gives a positive return on investment.  Fig 4  shows the relationships between supply and demand for uranium [10].  Supply is controlled by the cost of exploration, mining operations  and natU extraction;  demand is controlled by the number of users who want it.  Two demand curves are shown, one is for the power generation industry and the other is said to include naval uses.

Total world-wide demand is about 71 kt / year.  This is from Fig 4.  It is interesting to note that supply only contributed 54 kt of natU.  Currently, the difference is being made up from military stock, mostly from the former Soviet Union states that are dismantling their atomic weapons supply and selling a mix of  plutonium and depleted uranium left over from the enrichment process.  The result is MOX, discussed in Decisions-4 and usable in most light water reactors.  International agreements on this end in 2013, and mining must make up the natU difference.


Fig 5 – Uranium prices years starting 2006

Uranium prices rose as the reality of disappearing cheap oil began to impinge, Fig 5 show spot price history.  The power industry renewed exploration and began application of truly dirty mining techniques.  This price trend has an impact on estimates of uranium reserves.

World proven reserves amount to 5.4 Mt of uranium.  [11]   This is the Reasonably Assured Resource (RAR) value for when the selling  prices is US$ 130/kg.  The situation is slightly different than that for petroleum, where the oil is found only in well defined pools.  Since uranium exists in almost all types of rock throughout the world, the RAR always must be tied to the selling price, not to an estimate of a finite number of pools.  In 2007, we saw estimates of 6.3 Mt, but check the US$/kg levels of Fig 5.

There is a twist, however, that I have not seen discussed. Several things must be happening right now. Refer to the discussion between Figs 3 and 4, above. As easy petroleum disappears and the price of oil rises and the “front end” enrichment costs of extraction must be going up fast. This would have a negative effect, but uranium produces much more power kg for kg than oil and the value of finished LEU will increase as the petroleum supply decreases. [Ok, here is what happens— At peak oil, we will not see a reduction in the barrels of oil available, but we will see the loss of ability to meet the relentless growth in energy demand. This will cause a premium to appear on electric power, one that uranium based sources almost certainly will try to fill, CO2 be damned.]

Estimates in Uranium Reservoir Lifetime – 76 years.

Assume that the enriched and processed uranium is once in a reactor. When fission products build high enough to poison the reactions, that fuel is removed and sent to sequestration, a limbo existence for the next million years.

Lifetime for reserve TU = (5.4 Mt available) / (71 kt/yr use) = 76 years. This is close to the  value given in the literature [12, 13].

Will this get bigger?  Yes, when prospecting efforts return to earlier levels.  There is a lower limit to usable ore concentration in the soil because extraction becomes too expensive and generates too much CO2 to justify.  There has been an interesting scientific study in Japan in filtering natU from sea water, but this is truly at the level of an enjoyable experiment or a very early prototype.   Sea water extraction may actually become feasible, but as with most things, once engineering realities hit, financial and environmental impacts will shoot up.  This has happened in almost all industries.  Atomic power, fusion power, superconductivity cables, fiber optic transmission lines, space exploration –   none of  the big technologies are home runs, they have all proven to be real  challenges.


[1]  This table combines data taken from a number of sources, including:,,,
• link for fuel equivalents removed by request  2014 Aug 25

There are a number of grades of coal and wood which makes any standardized comparison difficult.  Uranium energy might mean fissile 235U,  or any of the possible enrichment levels of LEU ( anywhere in 3 to5%).  It could mean the natural U, U3O8, (yellowcake) found in ore with reported energy content scaled for use in LEU.  Or it could be natural U with much larger energy reported as when used in a CANDU class reactor.

The last row in our Table 1 is for  light water reactor grade fuel enriched to 4.3% in 235U.  It makes sense to compare fuels to the natural uranium (natU or NU) which is the material that is mined, not LEU, since all reserves are listed for yellow cake mining and extraction. We are not really concerned with the reactor capability, just the natural uranium in ore.   Return

[2] EIA-nuclear.  EIA is the US governments source of energy-related data.  Return

[3]  Petroleum usage:   EIA-petroleumReturn

[4]  Coal usage:   EIA-Coal. Return

[5]  Steve Fetter, “How long with the world’s uranium supplies last” Scientific American, March 2009.  The version referred to here was the on-line response. published  January 26, 2009. Return

[6]  World usage in kWh energy units is from  EIA  but at a slightly lower value than in [5]. Return

[7]  Much of the data for this post comes from the World Nuclear Association, an activity coordinator with most of the nuclear-capable counties as members. WNA-Inf75.   List of WNA resources.   Return

[8]  WNA-inf75 .     Return

[9]  David Flemming’s Lean Guide, especially Sect. 2, p8.  Also see Oxford Research Group, Jan Willem Storm van Leeuwen’s Energy from Uranium Sect 4 for costs and later for ore grades.   Return

[10]  Supply and demand is in many sources. Here is source from World Nuclear Association WNA-inf23Return

[11]  Reservoir estimates at  U-mining.   The IAEA estimates that at the start of 2009, RAR = 6.3 million t for a price of US$300/kgReturn

[12]   The Nuclear Energy Agency (NEA) is a specialised agency within the Organisation for Economic Co-operation and Development (OECD), an intergovernmental organisation of industrialised countries, based in Paris, France.    The NEA publishes (for free download) the discussion, Nuclear Energy Today   The OECD’s “Red Book” may be the best source on the quantity and quality of the remaining uranium ore, and of future prospects for production. Return

[13]  A good, short source to see the issues: The Lean Guide to Nuclear Energy, by David Flemming.  Return

About LastTechAge

I am a physicist with years of work in fusion labs, industry labs, and teaching (physics and math). I have watched the tech scene, watched societal trends and am alarmed. My interest is to help us all improve or maintain that which we worked so hard to achieve.
This entry was posted in Technology and tagged , , , , , , , , , , , , , . Bookmark the permalink.

6 Responses to Is there enough Uranium ore to do the job?

  1. Chris Baird says:

    As the total oil reserves become depleted and assuming there still is no cheap solution in solar/wind/wave energy, don’t you think that the planet will be forced to use nuclear regardless of what accidents happen around the globe?


    • LastTechAge says:

      Yes, and, countries have announced plans to build pressurized water reactors to replace oil dependance. France has done a good job of this and Japan had a fleet of these things scheduled. Each vulnerable to unexpected events, each building up its own spent waste stock. But there is not enough U3O8 to supply full
      power needs for many countries. France alone, yes. Japan, yes. Indonesia, yes. but don’t boolean AND these country needs together. — Charles Armentrout


  2. Chuck Hilton says:

    At $160/kg (almost double today’s prices), uranium contributes 0.1 cents per kwh to the cost of a kwh electricity (based on your figures). At 10 times these prices, the cost is 1 ct/kwh. At about $1100/kg, an infinite supply of uranium can be recovered from the oceans. We would never run out.


    • LastTechAge says:

      I think you are discussing extracting U from the water, not mining the nodules. Really interesting idea. It would be interesting to see if cost and efficiency match expectations. Many bright new ideas look foolproof until tried. Implementation always uncover new issues that must be met. But — there is something very attractive about nearly unlimited sources. What are the actual cost of finished uranium pellets? Only trying it will show the answer..

      What if ocean-based resources do open door to economical long term power generation? We still need a way to provide intrinsically safe reactors with sane handling of spent fuel. Proposals/plans currently on the table do neither.


      • Chuck Hilton says:

        I am quoting others – $500/# of yellow cake, using expensive ion exchange designed to extract uranium in the presences of huge amounts of other salts. Whether this is actually achievable won’t be known until prices approached this level. However, many conventional sources become economical well before $200 or even $100/# becomes the expected price.

        Handling waste remains a problem due to our gutless wonders in Washington. Yucca Mountain would have contained all waste produced for thousands of years.

        Excluding the stupidity of locating key pumps below known tidal surge levels in Japan, we’ve had exactly one nuclear disaster in 60 years in a dangerously designed carbon moderated plant in Russia. (Three mile island released virtually no radiation off-site). New designs are much more fail safe, and future designs will be better.

        No better way of reducing CO2 emissions without doubling or tripling our cost of energy, yet. Solar is getting there (need $1/watt installed). Solar remains my best hope with many more nuclear reactors as a back up if solar doesn’t work out.


  3. LastTechAge says:

    Waste handling is a serious engineering challenge. Thousands of years sounds huge but this is what has been described as “lies of large numbers” in other posts. MOX has typical half life values in the range of thousands to 10s of thousands of years. Safe level of decay activity requires a judgement call … at least 10 half-lives (drop by 1000x) or 30 half lives (drop by a billiion x). Read our post on spent fuel .

    What is a safe storage location? What is a safe storage cask? Proposing storage for geological time frames is really dumb. (put yourself into the position of someone wandering unawares through a radiation site.) Stupid is the decision that sequestration is an option.

    Reactors should also be a concern to anyone who wants to think about it. The Space Shuttle was vetted at a fail per 2 or 3 thousand launches. It’s record was 1 in every 50. (Maybe NASA actually knew that launch 150 ± 10 was to be a disaster so scuttled the program, blaming the Admin? — that was my idea of a joke!) Gen-1 reactors were supposed safe for many thousands of reactor years. That we know of, there are been 4 core-melt incidents in the 60 years since our Atoms For Peace program. Actually if you count the incidents with breeders in England and France, the number goes up. Reactor safety at and our first post on reactors,

    Our position is that safe reactors really have been built, just not by Westinghouse or GE. New proposals for modular gas cooled designs continue the intrinsically safe proposals, but what NRC is agreeing to are the techno-fixes applied to techno-interlocks to keep a core from thermally running away as it wants to. Most sound nice, until you factor in multiple things going wrong together. This comment could go on and on. yes we could build high power density generation sites using fission designs, but the US almost certainly will not. I think I will stop here and go turn off the fusion reactor in my backyard – wait! that option was gutted in the early 80’s, wasn’t it? (see economics section of our MFTF post )


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.