William Hannum

Dr. Hannum retired after more than 40 years in nuclear power development, stretching from design and analysis of the Shippingport reactor to the Integral Fast Reactor.  He earned his BA in physics at Princeton and his MS and PhD in nuclear physics at Yale.  He has held key management positions with the U. S. Department of Energy (DOE),  in reactor physics , reactor safety, and as Deputy Manager of the Idaho Operations Office.  He served as Deputy Director General of the OECD Nuclear Energy Agency, Paris, France; Chairman of the TVA Nuclear Safety Review Boards, and Director of the West Valley (high level nuclear waste processing and D&D) Demonstration Project.  Dr. Hannum is a fellow of the American Nuclear Society, and has served as a consultant to the National Academy of Engineering on nuclear proliferation issues.

Canadian Nuclear Society
29th Annual Conference
2 June 2008


Sensible recycling of used nuclear fuel will allow nuclear power to satisfy the early dream of environmentally responsible, essentially unlimited energy at a reasonable cost. This will require a multiple-pass nuclear fuel cycle.  Technologies for recycling used nuclear fuel are available that will resolve the most challenging nuclear waste issues and will significantly simplify the task of controlling the potential for weapons proliferation.   A major effort is needed to build prototype facilities for processing used fuel from today’s nuclear power plants, to recover material for use in fast reactors.  As these technologies are being developed and implemented, many additional nuclear power plants based on today's single-pass nuclear fuel cycle will be needed to meet near term demands for energy.


This is an exciting time to be involved in the nuclear power business.  Existing nuclear power plants are operating very well.  They are largely paid off, and are running flat-out, minting money.  Generating companies can see the need for additional base-load capacity, and there are no real competitors to nuclear power to fill this need.  But there are doubts and challenges on the horizon.

I am delighted to have the opportunity to address this gathering on what those challenges are, and on ways to address them.


Let me begin with a few disclaimers and qualifiers:


  • I do not speak on behalf of any program or company.  The opinions expressed are my own;
  • My opinions do not necessarily coincide with those of the U. S., or of any other government or government agency;
  • I do not wish to promote a particular technology; only an approach.  I have full confidence that there are smarter people than I who will find a way, given the goal and freedom to achieve, to develop the practical details;
  • I am speaking of an advanced technology.  The need for advanced technologies must not be used as an excuse to defer addressing immediate problems.  Advanced reactors capable of efficiently recycling used nuclear fuel will be needed if nuclear power is to grow significantly in the long run; but such technologies will not be developed if nuclear power does not grow significantly in the short run.  This is a chicken and egg issue.

The current situation

For this audience, I don’t think I need to belabor the need for additional energy.  One simple table should suffice:

Table 1

Population        Relative per capita energy use

U. S. + Canada              0.3 B                10

India + China                 2.5 B                  1

If, in the next decade of so, India, China, and other areas such as Africa, increase their energy use to, say, 20% of that of the U. S. and Canada, world energy demand will effectively double, regardless of what happens in the U. S. and Canada.

On the supply side, it is clear that nuclear power will be needed to supply much of this demand.

Coal is facing stringent new and expensive emission restrictions, potentially even CO2 taxes; natural gas is now prohibitively expensive (and wasteful) as a fuel for base load; there are few, if any, good, new sites left for major hydro installations.

Windmills, moonshine whiskey (corn-based ethanol) to fuel our cars, and coal plants with upside-down smoke stacks (to sequester CO2) may help, but they will not do the job.  The stability of our civilization requires massive expansion in the use of nuclear power.

As for nuclear power, several new reactor designs are available that offer simplifications and economies, with greatly improved safety.  Some of these designs already have licensing approval by regulators in the U. S. and elsewhere.  Even some politicians are speaking favorably of nuclear power.

So, if the situation is so positive for nuclear power, why do we have to go halfway around the world to see construction cranes actually building new nuclear power plants?

Aside from the lack of a competent energy plan, and the fact that utilities do not seem to want to take action on their long-term plans, there are two serious, legitimate concerns that need to be addressed:

  • Nuclear waste, and
  • The threat of nuclear terrorism.

No single technology will solve all our energy problems, but properly managed, recycling will resolve the nuclear waste issue, thereby removing the first of these impediments to nuclear energy becoming a major, potentially dominant resource in meeting our energy needs. 

The threat of nuclear terrorism will not be solved by burying our head in the sand, and burying partially-used nuclear fuel in a hole in the ground.  Recycling, if properly implemented, is one action that will significantly reduce the threat of nuclear terrorism.

Today’s Nuclear Power

I don’t think I need explain for this audience that using a major part of the energy content of mined uranium requires the use of fast reactors, where all of the uranium and other transuranic elements contribute to the neutron balance. And recycling is required to keep the fission products from poisoning the neutron balance.

We know how to build fast reactors; several are currently running very successfully, producing significant quantities of electricity.  But these reactors are not coupled to an appropriate recycle technology. The few fast reactors that are currently operating are orphans, whose only fuel comes from weapons programs and whose wastes are as bad as that from single-pass reactors.

Thus, today’s nuclear power is mainly characterized by single-pass (once-through) nuclear power plants that use less than 1% of the energy content of the ore mined.  The remaining 99% is considered to be an untreatable, hazardous waste that is strangling the prospects for nuclear power.  CANDU reactors do somewhat better than plants using enriched uranium, and the French recover and recycle some plutonium, but these approaches are still woefully inefficient ways to use uranium.

This does not need to be the case.

When I was in high school, I studied psychology by reading that classical expert: Dear Abby, and her Advice to the Lovelorn.  I recall two specific bits of wisdom from that:

On one occasion, a writer said: I’ve been going with this girl for 3 years, and I can’t get her to say “yes.”  What should I do?
Abby responded: What’s the question?

Another time, a writer said: I’ve tried flowers, candy, and even liquor, and I can’t get her to say “Yes.”  What should I try next?
Abby’s response: Try another girl.
Applying that wisdom to our situation, we see the following:

What IS the question?

It has been said: It is easier to reach your goal if you know what your goal is.

To use nuclear power to address the world’s energy needs, we need to effectively use our uranium resources.  To do that, we need to address two challenges:

  • How do we transform the used fuel from today’s single-pass reactors into fuel for fast reactors?, and;
  • How do we recycle fast reactor fuel so as to effectively utilize the energy content of the original ore?

For 40 years, we’ve been trying to find a way to adapt the PUREX process to the task of separating material from used single-pass reactor fuel to fuel fast reactors.  PUREX is the process that was developed to separate weapons usable material from used nuclear fuel.  Envisioning this as the foundation for an enduring nuclear energy economy, the PUREX technology was made publicly available, and a demonstration commercial reprocessing facility (West Valley, New York) was built and operated.  This strategy could have led us to a fast-breeder based economy if the world were a more peaceable place.  But it is not.

This strategy fell apart when it was recognized that it would lead to international commerce in plutonium, as other countries built PUREX type facilities.  With plutonium widely available around the world, it would be difficult to prevent the uncontrolled spread of nuclear weapons. 

This has become almost a moot point today.  There are now vast quantities of plutonium and enriched uranium spread around the world.  The threat of nuclear proliferation, and even nuclear terrorism, has grown dramatically with the spread of centrifuge technology and equipment.  Further spread of PUREX type facilities would be one more source of nuclear weapons material.

Are there alternate ways of recycling used fuel for use in fast reactors, in the light of concerns over nuclear waste and the threat of nuclear weapons proliferation and nuclear terrorism?

Try another process:

The desired process should take the used fuel, recover all of the usable energy content, and leave a waste stream that can be dealt with comfortably.  Chemists are clever, and have come up with several processes that will accomplish this.  I will speak briefly of two such processes: a used-fuel separations technology, and a fast reactor recycle technology.

After introducing these technologies, I will spend the remainder of my time addressing the implications that large scale recycle will have on nuclear power and its waste, and on the threat of nuclear proliferation.

You will note that I distinguish between a recycle technology and reprocessing, which is the term traditionally applied to the PUREX process.  You will also note that I speak of used nuclear fuel.  Used nuclear fuel should never be referred to as waste. 

I will then conclude with an outline of what will be needed to pursue such a strategy.

Used Fuel Separations

The first process I will discuss goes under the name UREX.  This is an adaptation of the PUREX process.  It focuses on extracting a “clean” waste stream that contains no plutonium or other transuranics and no uranium.  This is the inverse of the traditional PUREX process.  As with PUREX, the used fuel is dissolved in nitric acid.  Fission products and then uranium are chemically extracted.  Then the residue is reduced to metallic form, which, after blending back an appropriate amount of uranium, is used to make fast-reactor fuel.

The product is relatively clean, radiologically, so it is envisioned that certain fission products would be blended into the product to help protect it until it reaches a secure site and is used in a fast reactor.

Fast Reactor Recycling

The other technology I will speak of is called pyro-processing.  In principle, this is a reasonably straightforward process.  A batch of used fuel is chopped and placed in a bath of chloride salts.  An electrical current is then passed from the used fuel to a collector.  The salt (KCl-LiCl) is such that plutonium and the other transuranic elements (Np, Cm, Am) are the most efficiently transferred.  The process can be run so that essentially no plutonium or other transuranics remain in the salt.  Uranium is partially transferred, partially left in the salt.  A significant fraction of the fission products carry over or are encapsulated as the plutonium collects; the remainder is left in the salt. 

After extracting the plutonium and other transuranics, the salt is cleaned by first collecting the remaining uranium.  Then the fission products are chemically removed, and the salt is returned for the next cycle.

Thus, the products are:

  • Fission products with no uranium or transuranics;
  • Clean uranium; and
  • A melange, containing all the plutonium and other transuranics, some uranium, and a fraction of the fission products.

This third mixture is not pretty, but it is an ideal fuel for recycling back into a fast reactor.

Both of these processes have the distinct advantage that the waste output is naturally segregated into several distinct streams. Let me remind you that the hazard of nuclear waste falls into three time categories.  For the short time period (months to a few years), used fuel is both thermally hot and emits intense radiation.  This is a storage period.  This is the case whether the fuel is to be recycled or treated as a waste. 

The second (medium) period is dominated by Cs and Sr.  These materials emit significant radiation (they are used in commercial processes, such as food irradiation and well-logging).  They are also soluble, so if they are considered waste, they must be contained until their radiation dies away - a few hundred years.  Several waste forms, including borosilicate glass and a ceramic waste form known as syn-rock, have been shown to be more than adequate for this.  Hopefully, at some point these materials can be diverted to productive use, such as food preservation.

The waste problem with the single-pass fuel cycle, where partially-used fuel is classed as waste without any processing, is the presence of plutonium and other transuranics in the waste stream that remain radioactive for very long times.  If either the separation or the recycle technology discussed above is employed, this third, very long-term part of the nuclear waste problem goes away, since all of these materials are sent to the fast reactor as fuel. 

With complete recycle, the nuclear waste problem is reduced to a few hundred years, where proven containment strategies are readily available.

A mature recycle economy

Before getting into proliferation, I need to sketch what a mature recycle economy would look like.

Fast reactors will initially be fueled with enriched uranium and plutonium that has been declared excess from nuclear weapons programs.  There are hundreds of tonnes of such material, much of which is stored in secure bunkers, but a considerable amount is still in unaccounted stores and in scrap.  Existing stocks of used fuel, and additional quantities generated by current and planned single-pass nuclear power plants will be processed, with the product used to fuel additional fast reactors.  The fleet of fast reactors, once started, will operate on a closed fuel cycle, where the site receives only depleted, natural, or recycled uranium, or material to be incinerated.  The products shipped from the fast-reactor complex will be electricity, processed nuclear waste that will decay to harmless levels in a few hundred years, and very minor quantities of material requiring special handling.  Later, the fast reactor cycle can be modified to provide initial fueling for additional fast reactors as needed.

Proliferation liabilities

In discussing the potential impact of recycling on proliferation risks, let me first address the question: Can the separations and recycle technologies discussed above be used to produce nuclear weapons materials?

For the UREX process, an additional step would be necessary—one that separates plutonium from the other transuranics.  Either that, or the separations plant could be fed with specially irradiated (i.e., very briefly used) fuel, so that the plutonium in the feed would be uncontaminated with heavier isotopes.  The system would then have to be operated in a totally off-normal fashion.  It would seem that any modest form of surveillance would be able to detect such a diversion. 

Since UREX is a continuous process, a diversion would take a considerable time.  It is unlikely that there would be more than a few such plants, and it is expected that these plants will be located in advanced industrial countries, where alternative means of acquiring nuclear weapons materials would be easier than bypassing surveillance.  The safeguards policies and procedures would be based on, but simpler than those of weapons-based PUREX plants.

The process could be subverted, but this would be difficult, time-consuming, and easily detected.  Safeguards processes are well developed.

The pyro process has two additional safeguards features.  Being a small-scale batch process, it is likely that the recycle plant would be co-located with the fast reactors it services, so any off-site shipment would automatically be suspect.  Even if specially selected used fuel, high in plutonium and low in higher isotopes, were fed to the process, it would still not yield weapons usable materials, because the process does not cleanly separate plutonium from uranium.  Further, all operations will necessarily be conducted in highly shielded facilities, and any materials removed from the facility would have readily identifiable radiation characteristics.  The subsequent process to separate out plutonium would be a totally foreign process.

The process could be diverted to making feed material for a PUREX type process, but would not itself produce weapons-usable materials.

In either case, strict safeguards and accountancy procedures will be required, but there is no reason to suspect that safeguards would not be technically straightforward.

These modest additional safeguard concerns must be weighed against the dramatic reductions in the risk of uncontrolled spread of nuclear weapons that will result from an effective recycle program.

Proliferation benefits

The first substantial impact on the threat of nuclear terrorism or weapons proliferation will be to provide an efficient means of “denaturing” and incinerating excess nuclear weapons materials.  Fast reactors can do this some five times faster than the current generation of reactors, with far less complication.  And the used fuel, after denaturing, will not become another complicated special waste.  In a fast reactor, even much of the scrap from the nuclear weapons programs can be used as fuel.

The system will provide a market for used nuclear fuel, rather than leaving it as a temptation.  Perhaps as important as anything else, with a market, there will be a credible basis for much tighter inventory control on all nuclear materials than there is at present.

  • When the fast-reactor system is mature, there will be no inactive inventory of plutonium anywhere.
  • There will be no plutonium mine, which some people postulate could be a hazard in the distant future.

The system will minimize the need for enriching uranium, simplifying the problem of safeguarding these facilities.  Any effort to construct a uranium enrichment facility or a PUREX type of facility would be prima-facie evidence of a nuclear weapons program.

A brief word is in order here on the general question of nuclear terrorism.  While the public perception is that plutonium is the biggest concern, this is far from the truth.  While plutonium makes the best bombs, a uranium weapon is much easier to construct and easier to hide.  Reactor-grade plutonium would make a far less threatening weapon than would uranium.  Since the advent of centrifuge enrichment and the A. Q. Kahn network for obtaining this technology, enriched uranium is far more available than plutonium.

How do we get to a mature recycle economy?

There are five things that are needed to pursue this path:

1.    Construct a substantial number of new evolutionary-design nuclear power plants.  This will accomplish the following:

  • Address immediate power needs
  • Rebuild the infrastructure
  • Put money into the system

So far, there are very promising plans, but little action in the United States or Canada.

2.    Construct a large-scale separations plant to recover recyclable materials from current used reactor fuel.

  • Resolves the politically sensitive nuclear waste issue.
  • Provides a reliable supply of fuel for an expanding fleet of fast reactors

Toward this end, a program called the Global Nuclear Energy Partnership (GNEP) includes a commercial-scale demonstration plant to separate long-lived radioactive materials from current “nuclear wastes,” and to prepare them for incineration. 

GNEP provides a credible and sensible justification for building such a facility.  Again, the plans look good, but action is needed.

3.    Construct a prototype fast reactor, fueled with surplus weapons materials.

  • The technology is available, as are credible designs.  The economics and reliability of such a plant need to be demonstrated.

The GNEP program has solicited proposals for the construction of a fast reactor to consume the long lived radioactive materials recovered from used fuel in the separations plant (Item #2) . 

The design of the reactor should recognize its much broader potential, beyond the political objective of incinerating troublesome materials from current used nuclear fuel.

4.    Design and construct a demonstration facility for recycling fast-reactor fuel.

  • This technology has been demonstrated only at the feasibility stage.  Detailed designs are necessary to establish the economics and operability of such a facility.
  • This facility will be necessary, even if the fast reactor is no more than an incinerator.

5.    Revise international non-proliferation and safeguards agreements to reflect recycle technologies and to discourage new enrichment facilities.

  • While primarily a political issue, not a technical problem, this may be the most challenging part.  But it will be even more critical if we do not proceed with recycling.
  • GNEP has proposed a scheme whereby selected, trustworthy countries or some international organization would provide enrichment services for those building single-pass nuclear power, with all used fuel returned to the enriching country.  For smaller facilities, this service would be provided via sealed, plug-in reactor cores.
  • A program of full recycle, with appropriate transparency, would be a reasonable complement to such a program.

The GNEP program addresses the technical aspects, but this must be accompanied by a major diplomatic effort, that begin with an acknowledgment that the current Non-proliferation Treaty (NPT) has no enforcement provisions, and is quite ill-equipped to stem the spread of centrifuge enrichment.


The prospects for expanded use of nuclear power are the best they have been in over 40 years.

There are, however, serious impediments to expanded use of the current types of nuclear power plants: nuclear waste and the threat of proliferation.

An intelligent program of recycle is feasible.  The basic technologies are available.

Recycle offers a secure path to addressing many of the short- and long-term challenges to effective use of nuclear power.  Such a program will effectively address the nuclear waste issue, and will greatly reduce the threat of nuclear terrorism and weapons proliferation.

The biggest threat presented by recycle technologies is that they could be used as an excuse to defer or interfere with other necessary programs, such as constructing new evolutionary plants or GNEP.

With recycle, nuclear power offers an essentially unlimited energy resource with which to power a growing world economy.