The DIY Garage Nuke Method – Or how to make a bomb without really trying…

Disclaimers and Caveats

First off, we need some disclaimers and caveats.

I’ve been thinking about this since about 1985 or so. I’ve sat on ALL of it for about 30 years. I’ve watched as bits and pieces, dribs and drabs have entered the public domain. At this point, substantially all of it is easily found in just a few hours of web searching, so it isn’t like I’m giving away a bunch of deep dark secrets.

Second, nuclear engineering is not hard at all, conceptually. It is terrifically hard in the details. The details matter. So giving a conceptual guide does not really provide “aid and comfort” to anyone. They are as likely to kill themselves as anyone else if they don’t already know the details. (Little things are left out, like, oh, how many gamma of high energy you are likely to absorb if you actually try any of this and just how fast your flesh will melt…) So please, no carping about “aid and comfort”, OK?

Finally, I’m a complete amateur at this. Now on one hand, it is startling to me that, as an amateur, I have a decent clue on how to go about this. OTOH, anything I say here could easily spell doom for anyone foolish enough to think this will work. Frankly, I haven’t a clue if it will work. (Well, I have a clue, just not a very good one, and it might well be a suicide pill…)

Political Dimension

Over the years, the nuclear club has grown to include folks like Red China, Pakistan, North Korea and soon Iran. These are among the least morals bound nations in the world and they already have “the bomb”. Folks who are relatively moral have also fallen into ‘less moral’ ways. These include the present Russia. Finally, we have the relatively moral countries of France, UK, USA, and Israel.

This excludes the “could have them if we wanted in no time at all” and the “had them before but probably don’t now” or the “NATO owns them be we have them, tee hee” countries of: Germany, Japan, South Korea, Belarus, Kazakhstan, Ukraine, South Africa, Belgium, Turkey, Netherlands, and even Italy. “How to make a Big Boom!” is one of the worst kept ‘secrets’ in the world. We largely depend on morals and hard to make SNM (Special Nuclear Materials that I like to call “boom stuff”) to keep things from going to hell in a hand basket “right quick”.

So, frankly, I’m not particularly bought into the notion that “security by obscurity” is going to work any longer. IMHO, just about ANY nuclear engineer can do a credible job of making a big-badda-boom and certainly far better than me. So, IMHO, at most I can “spill the beans” about what everybody in the “community” already knows.

That means I’m not particularly worried that anything I’m about to say:

a) Works.
b) Won’t get you killed.
c) Isn’t already widely known.
d) Isn’t already known to be the wrong path.
e) Is of little use to anyone, especially those with a desire to kill me.

In short, anyone who really wants to get The Bomb will be showing up in Pakistan or North Korea with a couple of $Billion and walking away with a working device, not reading some blog by a guy who has only looked at this from the outside for 40 years.

The Tech

First up, we need to state the goals. I’ve mostly been interested in two things. Smallest possible reactor that I could actually make if I really wanted to make one (and didn’t mind a visit from people with badges, uniforms, guns, and maybe even padded cells. I’d hope they would recognize my genius and offer me a job, but that isn’t likely. Why is left as an exercise for the student ;-)

I’ve only secondarily been interested in “path to a Big-Badda-Boom” and mostly as an intellectual exercise in information theory. (Hint: Look where the silence is greatest… the “negative space” tells a great deal.) As a minor point I’ve wondred “why doesn’t FOO look at BAR?” and found my own answers.

I’ve often said that I thought there was a “back door” to SNM, but rarely given hints to what it might be. As that “back door” is now being openly discussed, and it already proven to work (via the USA Teapot Met test and the Indian U233 bomb) it seems a bit daft to keep silent about that which is openly acknowledged. Besides, that “back door” comes with some pretty significant personal death probabilities from gamma rays and makes a device that gets ever more “toasty” over time while emitting a gamma signature that makes it pretty easy to spot. So at most the game moves from “ignorance is bliss” to “detection and intervention” wins.

So what is this hypothetical “back door”? U233. India made a bomb from it. USA tested a mix material bomb with it (and I’ve seen implied further tests). It is about as good as Plutonium (in some ways a bit better, in contaminant radiation a bit worse) and as Taylor (our best boutique bomb maker) said: There is good plutonium for bombs, and there is better Pu for bombs, but there is no bad Pu for bombs. (As a remembered paraphrase. For actual quote, read John Mcfee “The Curve of Binding Energy”. A great read and to some extent a biography of Taylor. )

Fissile material

In 1946 the public first became informed of U-233 bred from thorium as “a third available source of nuclear energy and atom bombs” (in addition to U-235 and Pu-239), following a United Nations report and a speech by Glenn T. Seaborg.

The United States produced, over the course of the Cold War, approximately 2 metric tons of uranium-233, in varying levels of chemical and isotopic purity. These were produced at the Hanford Site and Savannah River Site in reactors that were designed for the production of plutonium-239. Historical production costs, estimated from the costs of plutonium production, were 2–4 million USD/kg. There are few reactors remaining in the world with significant capabilities to produce more uranium-233.

I’d quibble over that line about “few reactors remaining” as the CANDU reactor can do it fairly easily. Just shove Thorium through one of the fuel pipes. The CANDU can run on Uranium or Thorium and in a pinch can have some Thorium run through relatively fast as a U-233 breeder operation without having so long an exposure as to build up too many unwanted isotopes. But maybe they see “significant capabilities” as being in the tons range, not the ‘couple of big bangs’ range…

Nuclear fuel

Uranium-233 has been used as a fuel in several different reactor types, and is proposed as a fuel for several new designs (see Thorium fuel cycle), all of which breed it from thorium. Uranium-233 can be bred in either fast reactors or thermal reactors, unlike the uranium-238-based fuel cycles which require the superior neutron economy of a fast reactor in order to breed plutonium, that is, to produce more fissile material than is consumed.
The long-term strategy of the nuclear power program of India, which has substantial thorium reserves, is to move to a nuclear program breeding uranium-233 from thorium feedstock.

I think you can see that this is largely well discussed and “old news” to anyone in the industry. So anything I can say is at most letting the “unwashed” in on the story.

U238 is the usual and most common form of Uranium. U235 is the usual “boom stuff” of SNM (for Uranium bombs). Little mentioned is U233 that is an even better “boom stuff” and much like Pu in that regard. U233 is easily made from Thorium. In fact, since Thorium is not fissile, all Th reactors pretty much depend on turning Th into U233 that is then fissile in order to work. Mostly folks expect that the “other isotopes” that make the whole process rather incredibly radioactive will make diversion of U233 into bombs “beyond the pale”. I suspect that official “guidance” away from Th and into U reactors was largely to divert attention from that simple fact / process.

So, with that in mind, the question becomes: How to make U233 without too much U232 (that makes it very “hot” with high energy gammas) and how to extract that easily from the process?

Normal power reactors are all about the opposite. How to keep the fuel so “dirty” that it is impossible to make a bomb (as attempts at explosive “assembly” will squib out first due to excess neutrons) or make it relatively deadly to the bomb maker (via so much ambient radiation that attempting to make the device kills you or lights up the whole place with a beacon of radiation easily seen from afar… while it cooks you…)

IMHO, this is why we have not had Throrium MSR (Molten Salt Reactors) commercialized to date. (We had them running decades ago…) Folks realized that the fuel would have the intermediate Protactinium-233 and that maybe it could be removed from the reactor, and then when it converts to U-233, the product is pure enough. There are many plans for this, widely discussed, so not a big deal. Most of them need some fairly complicated machinery and it’s all theoretical (at least until China gets their reactors running) so nobody cares much.

So the question becomes: How do you make relatively pure U-233 without too much U-232 in a not too complicated process? Preferably using chemical separation instead of things like isotopic centrifuge separation.

Boot Strap Reactor

First off, you need some way to get some more concentrated nuclear fuel than raw uranium deposits. Yes, many reactors can use un-enriched Uranium (such as the Canadian CANDU, that was suppressed to some extent by the USA in an attempt to prevent what actually happened. India used a CANDU like design to make the SNM for their “devices”…) So we could just go buy / build a CANDU like reactor. But that is big, and expensive, and takes tons of Heavy Water.

While deuterium oxide is readily available and not very expensive ( I have about 3 grams of it in a bottle somewhere… a gift from an engineer friend) it draws attention when you start buying it by the rail car. So is there any smaller, easier way to make a natural uranium fueled reactor with a lot less heavy water?


The “homogeneous reactor”.

In many ways, if you just look at ‘the path not taken’ through the history of the US nuclear plan, you find it littered with the easier means to an end. I’m tempted to say that it looks like every decision was made based on “hide what is easy, show the hard expensive path as working” so as to mislead folks for generations. If that was the case: Good job! It worked!. It also leaves a clear trail of crumbs for finding the “easy way” by just stepping off that path and exploring the “negative space”. (Negative Space Analysis is one of my best and favorite tools. I don’t know if it has ever been codified as a formal thing, but I’ve used it for decades to good effect. Look for what “ought to be but is not” and go there…)

Basically, you blend the U as a salt with water as the coolant and moderator and put it in a bucket. Yup, that simple. (Staying alive is left as an exercise for the student ;-)

Aqueous homogeneous reactors (AHR) are a type of nuclear reactor in which soluble nuclear salts (usually uranium sulfate or uranium nitrate) are dissolved in water. The fuel is mixed with the coolant and the moderator, thus the name “homogeneous” (‘of the same physical state’) The water can be either heavy water or ordinary (light) water, both of which need to be very pure.

A heavy water aqueous homogeneous reactor can achieve criticality (turn on) with natural uranium dissolved as uranium sulfate. Thus, no enriched uranium is needed for this reactor.

The heavy water versions have the lowest specific fuel requirements (least amount of nuclear fuel is required to start them). Even in light water versions less than 1 pound (454 grams) of plutonium-239 or uranium-233 is needed for operation. Neutron economy in the heavy water versions is the highest of all reactor designs.

Their self-controlling features and ability to handle very large increases in reactivity make them unique among reactors, and possibly safest. At Santa Susana, California, Atomics International performed a series of tests titled The Kinetic Energy Experiments. In the late 1940s, control rods were loaded on springs and then flung out of the reactor in milliseconds. Reactor power shot up from ~100 watts to over ~1,000,000 watts with no problems observed.

Aqueous homogeneous reactors were sometimes called “water boilers” (not to be confused with boiling water reactors), as the water inside seems to boil, but in fact this bubbling is due to the production of hydrogen and oxygen as radiation and fission particles dissociate the water into its constituent gases. AHRs were widely used as research reactors as they are self-controlling, have very high neutron fluxes, and were easy to manage. As of April 2006, only five AHRs were operating according to the IAEA Research Reactor database.

Corrosion problems associated with sulfate base solutions limited their application as breeders of uranium-233 fuels from thorium. Current designs use nitric acid base solutions (e.g. uranyl nitrate) eliminating most of these problems in stainless steels.

Here, on one small quote, we have all you need to get started. Use Uranyl Nitrate for longer term runs. But use Uranyl Sulphate for the first start up run to get enough U-233 to make the light water Uranyl Nitrate one run longer term. It is largely self controlling, needs little care in the operation, and works on very small quantities of SNM (or none for a U-SO4 approach).

Run your plain water U-sulfate reactor until you get a couple of pounds of U-233 in a breeder blanket, then swap to a light water version from that point on. Eventually collect enough U-233 (via a relatively clean method below) until you can not only run your breeder, but make a pile of “boom stuff” too.

Assemble and test. (From a distance, please, and you DID wear your lead underwear during this process, right? ;-)

Please note that I’m assuming it is required to use U-sulphate in the heavy water version, but that isn’t clear. It might well be that any of the salts known to work can be used in both light and heavy water versions. IF that is true, start with the U-nitrate from the get-go.

Some Sizing

OK, the Wiki has the idea, but is this thing REALLY small enough to fit in the garage or back yard? (Be aware that various programs look for odd gamma signatures, so you ought to expect a visit form the nice men with guns and badges even if done out of common view…)

From that same Wiki:

Enrico Fermi advocated construction at Los Alamos of what was to become the world’s third reactor, the first homogeneous liquid-fuel reactor, and the first reactor to be fueled by uranium enriched in uranium-235. Eventually three versions were built, all based on the same concept. For security purposes these reactors were given the code name “water boilers”. The name was appropriate because in the higher power versions the fuel solution appeared to boil as hydrogen and oxygen bubbles were formed through decomposition of the water solvent by the energetic fission products.

The reactor was called LOPO (for low power) because its power output was virtually zero. LOPO served the purposes for which it had been intended: determination of the critical mass of a simple fuel configuration and testing of a new reactor concept. LOPO achieved criticality, in May 1944 after one final addition of enriched uranium. Enrico Fermi himself was at the controls. LOPO was dismantled to make way for a second Water Boiler that could be operated at power levels up to 5.5 kilowatts.

Named HYPO (for high power), this version used solution of uranyl nitrate as fuel whereas the earlier device had used enriched uranyl sulfate. This reactor became operative in December 1944. Many of the key neutron measurements needed in the design of the early atomic bombs were made with HYPO. By 1950 higher neutron fluxes were desirable, consequently, extensive modifications were made to HYPO to permit operation at power levels up to 35 kilowatts this reactor was, of course, named SUPO. SUPO was operated almost daily until its deactivation in 1974.

In 1952, two sets of critical experiments with heavy water solutions of enriched uranium as uranyl fluoride were carried out at Los Alamos to support an idea of Edward Teller about weapon design. By the time the experiments were completed, Teller had lost interest, however the results were then applied to improve the earlier reactors. In one set of experiments the solution was in 25-and-30-inch-diameter (640 and 760 mm) tanks without a surrounding reflector. Solution heights were adjusted to criticality with D2O solutions at D/235U atomic ratios of 1:230 and 1:419 in the smaller tank and 1:856 to 1:2081 in the larger tank. In the other set of experiments solution spheres were centered in a 35-inch-diameter (890 mm) spherical container into which D2O was pumped from a reservoir at the base. Criticality was attained in six solution spheres from 13.5- to 18.5-inch diameter at D/235U atomic ratios from 1:34 to 1:431. On completion of the experiment that equipment too was retired.

Got that? We’re working with spheres of 13.5 inch to 30 inch diameter. Something you could put in a small truck and drive around. Easily fits in a garage or back yard. “Nuke in a jar” sized. Adding a neutron reflector helps make it smaller. What’s a reflector?

A neutron reflector is any material that reflects neutrons. This refers to elastic scattering rather than to a specular reflection. The material may be graphite, beryllium, steel, tungsten carbide, or other materials. A neutron reflector can make an otherwise subcritical mass of fissile material critical, or increase the amount of nuclear fission that a critical or supercritical mass will undergo. An example of this is the Demon Core, a subcritical plutonium pit that went critical in two separate fatal incidents when the pit’s surface was momentarily surrounded by too much neutron reflective material.

While buying a 30 inch (or even a 24 inch) “beryllium sphere” (shades of Galaxy Quest ;-) might draw attention, I doubt that steel would even be noticed. Oh, and plain old water makes a decent reflector, so if you have a swimming pool, just put the sphere in the middle of it…

We will see this again in device design:

A similar envelope can be used to reduce the critical size of a nuclear weapon, but here the envelope has an additional role: its very inertia delays the expansion of the reacting material. For this reason such an envelope is often called a tamper. The weapon tends to fly to bits as the reaction proceeds and this tends to stop the reaction, so the use of a tamper makes for a longer lasting, more energetic, and more efficient explosion. The most effective tamper is the one having the highest density; high tensile strength is unimportant because no material remains intact under the extreme pressures of a nuclear weapon. Coincidentally, high density materials are excellent neutron reflectors. This makes them doubly suitable for nuclear weapons. The first nuclear weapons used heavy uranium or tungsten carbide tamper-reflectors.

On the other hand, a heavy tamper necessitates a larger high explosive implosion system. The primary stage of a modern thermonuclear weapon may use a lightweight beryllium reflector, which is also transparent to X-rays when ionized, allowing the primary’s energy output to escape quickly to be used in compressing the secondary stage.

While the effect of a tamper is to increase efficiency, both by reflecting neutrons and by delaying the expansion of the bomb, the effect on the critical mass is not as great. The reason for this is that the process of reflection is time consuming.

Not mentioned is that putting your Boom Device in a swimming pool or the bottom of a boat under the lead “ballast” also does the job… So avoid living near the Potomac in DC…

For now, what we care about is that we can make that 30 inch sphere into a 1.x foot small or 2 foot sphere with some surrounding reflective stuff, be it steel, water, or beryllium.


The commonly used materials for neutron moderators and reflectors are light water (H2O), heavy water (D2O), graphite (C), zirconium hydride (ZrHx), and beryllium (Be). When utilizing such materials in nuclear reactors, fundamental properties such as thermal and mechanical properties should be understood. This chapter has provided an overview of the fundamental properties of beryllium and zirconium hydride required for neutron reflectors. The outline of zirconium hydride has described the influence of temperature and hydrogen concentration on the basic properties of the hydride.

Note that metal hydrides work and plain old water ain’t too bad…

So our first cut will be a U-sulphate based reactor, likely about a meter or 30 inches in diameter, using natural uranium, but with an added shell containing a Thorium “slurry” or similar aqueous formulation where we breed U-233 (and also acting a bit as a reflector and shielding) with an outer shell of a water jacket (or graphite / charcoal / zirconium hydride / Beryllium / {whatever} neutron reflector). This has to last long enough to breed a couple of pounds of U-233 in the blanket, or we get to re-make it enough times to get there. Whatever, it’s just time vs finesse.

I think you can already see where this is going. A “device” that needs no enrichment to run, using at most a 30 inch sphere of heavy water, and maybe a 60 inch total diameter with breeder blanket and water jacket. All up about a 5 foot sphere. Not exactly a large project.

Again I note for future improvement that it might be possible to use the other Uranium salts in place of the sulphate at this stage and skip some of the corrosion problems. I just don’t know the neutron flux vs. barns cross sections of the various salts well enough at the moment to say. For a natural U reactor, you need very good neutron economy. Once you can enrich the mix with U-233 then the economy isn’t as important.

Also note that putting a moderating water blanket in front of the breeder blanket will give slower neutrons and help make less U-232 contaminant.

Once you have the ‘couple of pounds’ of U-233 what’s next? (and it need not be low radiation. While that would make handling it easier, all you really need is the enrichment factor for the next step. A “crazy Ivan” or Jihadi can be found to help if needed, but I suspect that Lead Underwear and fast hands would be enough. Or care in how the breeder blanket is run.) You make the light water version of the same thing, using the breeder U-233 to ‘enrich’ the fuel of natural uranium.

This limits your initial need for heavy water to a fairly small quantity. A 2 foot sphere is Pi*1 volume or 3.1415 cubic feet. So at most about 3 cubic feet of heavy water. IIRC, it’s about 64 lbs / ft^3 for regular water, or about 190 to 200 lbs for regular water. Call that about 20 gallons to 25 gallons. Not exactly a large quantity, though it would likely require a team of buyers at a gallon or less each with a plausible cover story to pass notice. Still way below the trainload scale most folks think is an issue….

OK, so now we have done our first run with heavy water, have a breeder blanket of Th making U-233, and can move on to the enriched (with U-233) light water reactor from this point forward. At most, we’ve had a 30 gallon can of heavy water “tickle” of the monitoring systems, and the whole operation fits in a moving 18 wheeler anyway… so good luck finding where those shell companies each buying one gallon sent their goods… and where it is now.

The Breeder Running Stage

At this point we have a running nuclear breeder in a truck. It is using a light water design with a breeder blanket that needs Thorium (available in beach sand in the Carolinas or Florida and elsewhere) and light water, not much else. We might need a second truck as the ‘refinery’ to take the breeder blanket product and process it, but more on that later. It is about the timing of processes and the need to store ‘volume’ for a while… Bottom line is that we are breeding “Special Nuclear Material” and don’t need a farm of centrifuges to do it nor do we need an ongoing source of “suspicious” materials like heavy water.

We are over the hump of getting natural uranium isotopes to fission and into the realm of breeders and ‘enriched’ fuels without needing a single centrifuge or exotic method. Just chemical separation.

At this point the goal is maximal production of U-233 with minimal U-232 in it. Any fuel with too much U-232 goes into the reactor core as nitrate or phosphate. (Sidebar: Nitrate has fewer corrosion issues. I’ve seen a page saying a U-phosphate reactor was made in Los Alamos… then silence… so most likely the phosphate works best. In any case, the stainless steel corrosion issue is likely behind us now as is the heavy water phase. I’d do a lot more research on the U-phosphate cycle were I actually doing this instead of just making a blog posting out of it. The silence on that front is deafening…)

In any case, the reactor still fits in a mobile rig or easily can be put under a few dozen meters of dirt under a house somewhere…

We are, at this point, likely in the range of a 2 foot reactor with a 1 foot breeder / reflector shell; so the whole thing is about 4 foot diameter or less. Hell, it would fit in my office with a 2 foot concrete shield with room to squeeze past it to my desk…

At this point, it is largely just finding optimizing things and ways to better handle the chemistry. Likely Grad Student stuff.

Some Links and Background


In addition to the Aircraft Reactor Experiment, the Bulk Shielding
Reactor, and the Tower Shielding Facility built as part of its
Aircraft Nuclear Project for the Air Force, the Laboratory had
three other major reactor designs in progress during the mid-1950s:
its own new research reactor with a high neutron flux; a portable
package reactor for the Army; and the Aqueous Homogeneous Reactor,
which was unique because it combined fuel, moderator, and coolant
in a single solution (designed as one of five demonstration
reactors under AEC auspices).

Initial studies of homogeneous reactors took place toward the close
of World War II. It pained chemists to see precisely fabricated
solid-fuel elements of heterogeneous reactors eventually dissolved
in acids to remove fission products–the “ashes” of a nuclear
reaction. Chemical engineers hoped to design liquid-fuel reactors
that would dispense with the costly destruction and processing of
solid fuel elements. The formation of gas bubbles in liquid fuels
and the corrosive attack on materials, however, presented daunting
design and materials challenges.

With the help of experienced chemical engineers brought to the
Laboratory after its acquisition of the Y-12 laboratories, the
Laboratory proposed to address these design challenges. George
Felbeck, Union Carbide manager, encouraged their efforts. Rather
than await theoretical solutions, Laboratory staff attacked the
problems empirically by building a small, cheap experimental
homogeneous reactor model. Engineering and design studies began in
the Reactor Experimental Engineering Division under Charles
Winters, and in 1951 the effort formally became a project under
John Swartout and Samuel Beall.

This was the Laboratory’s first cross-divisional program. Swartout
provided program direction to groups assigned in the Chemistry,
Chemical Technology, Metallurgy, and Engineering divisions, while
Samuel Beall led construction and operations. Beecher Briggs headed
reactor design; Ted Welton, Milton Edlund, and William Breazeale
were in charge of reactor physics; Edward Bohlmann directed
corrosion testing; and Richard Lyon and Irving Spiewak performed
fluid flow studies and component development.

A homogeneous (liquid-fuel) reactor had two major advantages over
heterogeneous (solid-fuel and liquid-coolant) reactors. Its fuel
solution would circulate continuously between the reactor core and
a processing plant that would remove unwanted fissionable products.
Thus, unlike a solid-fuel reactor, a homogeneous reactor would not
have to be taken off-line periodically to discard spent fuel.
Equally important, a homogeneous reactor’s fuel and the solution in
which it was dissolved served as the source of power generation.
For this reason, a homogeneous reactor held the promise of
simplifying nuclear reactor designs.

A building to house the Homogeneous Reactor Experiment was
completed in March 1951. The first model to test the feasibility of
this reactor used uranyl sulfate fuel. After leaks were plugged in
the high-temperature piping system, the power test run began in
October 1952, and the design power level of one megawatt (MW) was
attained in February 1953. The reactor’s high-pressure steam
twirled a small turbine that generated 150 kilowatts (kW) of
electricity, an accomplishment that earned its operators the
honorary title “Oak Ridge Power Company.”

Marveling at the homogeneous reactor’s smooth responsiveness to
power demands, Weinberg found its initial operation thrilling.
“Charley Winters at the steam throttle did everything, and during
the course of the evening, we electroplated several medallions and
blew a steam whistle with atomic steam,” he exulted in a report to
Wigner, asking him to bring von Neumann to see it. Despite his
enthusiasm, Weinberg found AEC’s staff decidedly bearish on
homogeneous reactors and, in a letter to Wigner, he speculated that
the “boiler bandwagon has developed so much pressure that everyone
has climbed on it, pell mell.” Weinberg surmised that the AEC was
committed to development of solid-fuel reactors cooled with water
and Laboratory demonstrations of other reactor types–regardless of
their success–were not likely to alter its course.

Despite AEC preferences, the Laboratory dismantled its Homogeneous
Reactor Experiment in 1954 and obtained authority to build a large
pilot plant with “a two-region” core tank. The aim was not only to
produce economical electric power but also to irradiate a thorium
slurry blanket surrounding the reactor, thereby producing
fissionable uranium-233. If this pilot plant proved successful, the
Laboratory hoped to accomplish two major goals: to build a
full-scale homogeneous reactor as a thorium “breeder” and to supply
cheap electric power to the K-25 plant to enrich uranium.

Initial success stimulated international and private industrial
interest in homogeneous reactors,
and in 1955 Westinghouse
Corporation asked the Laboratory to study the feasibility of
building a full-scale homogeneous power breeder. British and Dutch
scientists studied similar reactors, and the Los Alamos Scientific
Laboratory built a high-temperature homogeneous reactor using
uranyl phosphate fluid fuel. If the Laboratory’s pilot plant
operated successfully, staff at Oak Ridge thought that homogeneous
reactors could become the most sought-after prototype in the
intense worldwide competition
to develop an efficient commercial
reactor. Proponents of solid-fuel reactors, the option of choice
for many in the AEC, would find themselves in the unenviable
position of playing catch-up. But this was not to be.

Section 6.0 Nuclear Materials

Nuclear Weapons Frequently Asked Questions

Version 2.18: 20 February 1999

This material may be excerpted, quoted, or distributed freely
provided that attribution to the author (Carey Sublette) and
document name (Nuclear Weapons Frequently Asked Questions) is
clearly preserved. I would prefer that the user also include the
URL of the source.

I think that covers the requested attribution an links…

There is a LOT more of value at that link than I will quote here.

6.2 Fissionable Materials

There are three isotopes known which are practical for use as fission explosives. These are U-235, Pu-239, and U-233. Of these only U-235 occurs in nature. Pu-239 and U-233 must be produced by bombarding other isotopes with neutrons. A third element, thorium (Th-232), can only undergo fast fission, but can also be used for breeding U-233. There are other elements that are also fissile but they have no practical significance for a variety of reasons. These elements are summarized in subsection 6.2.4.
[…] U-233

This fissile uranium isotope (half-life 162,000 years) is not found in nature. It is instead bred from thorium-232 in a manner similar to the production of Pu-239:

Th-232 + n -> Th-233
Th-233 -> (22.2 min, beta) -> Pa-233
Pa-233 -> (27.0 day, beta) -> U-233

A two-step side reaction chain also occurs during breeding leading to the production of U-232:

Th-232 + n -> Th-231 + 2n
Th-231 -> (25.5 hr, beta) -> Pa-231
Pa-231 + n -> Pa-232
Pa-232 -> (1.31 day, beta) -> U-232
The production of U-232 through this process depends on the presence of significant amounts of un-thermalized neutrons since the cross section of the initial n,2n reaction is small at thermal energies.

So thermalize those neutrons in a water blanket or other moderator before they get to the Th slurry… that keeps the U-232 down and makes the rest much easier. Note also the relative times. 22 minutes for Th-233 to Pa-233 and 25 HOURS for Th-231 to Pa-231. While not very efficient, if you can get a high neutron flux into the breeder blanket, you can “cook” Th fast, then extract it (maybe 2 hours? that’s about 6 half lives of the Th-233) and remove the Pa-233 before significant Th-231 converts to Pa-231. ( about 1/10 of a half life). Between moderated neutrons, Thorium low in the unwanted isotopes, and fast processing of the breeder blanket, I’m guessing that pure enough Pa-233 can be chemically extracted with low enough radiation from other Pa isotopes in the final product.

If significant amounts of the isotope Th-230 are present then U-232 production is augmented by the reaction:

Th-230 + n -> Th-231
which continues as before.
The presence of U-232 is important because of its decay chain:

U-232 -> (76 yr, alpha) -> Th-228
Th-228 -> (1.913 yr, alpha) -> Ra-224
Ra-224 -> (3.64 day, alpha & gamma) -> Rn-220
Rn-220 -> (55.6 sec, alpha) -> Po-216
Po-216 -> (0.155 sec, alpha) -> Pb-212
Pb-212 -> (10.64 hr, beta & gamma) -> Bi-212
Bi-212 -> (60.6 min, beta & gamma) -> Po-212
alpha & gamma) -> Tl-208
Po-212 -> (3×10^-7 sec, alpha) -> Pb-208 (stable)
Tl-208 -> (3.06 min, beta & gamma) -> Pb-208
The rapid decay sequence beginning with Ra-224 produces a large amount of energetic gamma rays. About 85% of this total gamma energy output is due to the last isotope in the sequence, thallium-208 which produces the most energetic gamma rays (up to 2.6 MeV). The amount of gamma radiation emitted is proportional to the amount of Th-228 present.

The buildup of U-232 as a contaminant is unavoidable during the production of U-233. This is similar to the plutonium isotope contamination problem discussed below in plutonium production, but occurs to a much smaller extent rate. The first (n,2n) reaction only occurs when neutrons with energies in excess of 6 MeV are encountered. Only a small percentage of fission neutrons are this energetic, and if the thorium breeding blanket is kept in a reactor region where it is only exposed to a well moderated neutron flux (i.e essentially no neutrons above the Th-232 fission threshold of 500 KeV) this reaction can be nearly eliminated. The second reaction proceeds very efficiently with thermalized neutrons however, and minimizing U-232 from this source requires choosing thorium that naturally has a low Th-230 concentration.

If the above precautions are followed weapons-grade U-233 can be produced with U-232 levels of around 5 parts per million (0.0005%). Above 50 ppm (0.005%) of U-232 is considered low grade.

In a commercial fuel cycle the build-up of U-232 is not really a disadvantage, and may even be desirable since it reduces the proliferation potential of the uranium. In a fuel economy where the fuel is reprocessed and recycled the U-232 level could build up to 1000 – 2000 ppm (0.1 – 0.2%). In a system that is specifically engineered to accumulate U-232 levels of 0.5-1.0% can be reached.

Over the first couple years after U-233 containing U-232 is processed, Th-228 builds up to a nearly constant level, balanced by its own decay. During this time the gamma emissions build up and then stabilize. Thus over a few years a fabricated mass of U-233 can build up significant gamma emissions. A 10 kg sphere of weapons grade U-233 (5 ppm U-232) could be expected to reach 11 millirem/hr at 1 meter after 1 month, 0.11 rem/hr after 1 year, and 0.20 rem/hr after 2 years. Glove-box handling of such components, as is typical of weapons assembly and disassembly work, would quickly create worker safety problems. An annual 5 rem exposure limit would be exceeded with less than 25 hours of assembly work if 2-year old U-233 were used. Even 1 month old material would require limiting assembly duties to less than 10 hours per week.

In a fully assembled weapon exposures would be reduced by absorption by the tamper, case, and other materials. In a modern light weight design this absorption would be unlikely to achieve more than a factor of 10 attenuation, making exposure to weapons assembled two years previously an occupational safety problem. The beryllium reflectors used in light weight weapons would also add to the background neutron level due to the Be-9 + gamma -> Be-8 + neutron reaction. The U-232 gammas also provide a distinctive signature that can be used to detect and track the weapons from a distance. The heavy tampers used in less sophisticated weapon designs can provide much high levels of attenuation – a factor of 100 or even 1000.

With deliberately denatured grades of U-233 produced by a thorium fuel cycle (0.5 – 1.0% U-232), very high gamma exposures would result. A 10 kg sphere of this material could be expected to reach 11 rem/hr at 1 meter after 1 month, 110 rem/hr after 1 year, and 200 rem/hr after 2 years. Handling and fabrication of such material would have to done remotely (this also true of fuel element fabrication) In an assembled weapon, even if a factor of 1000 attenuation is assumed, close contact of no more than 25 hours/year with such a weapon would be possible and remain within safety standards. This makes the diversion of such material for weapons use extremely undesirable.

The short half-life of U-232 also gives it very high alpha activity. Denatured U-233 containing 1% U-232 content has three times the alpha activity of weapon-grade plutonium, and a correspondingly higher radiotoxicity. This high alpha activity also gives rise to an even more serious neutron emission problem than the gamma/beryllium reaction mentioned above. Alpha particles interact with light element contaminants in the fissile material to produce neutrons. This process is a much less prolific generator of neutrons in uranium metal than the spontaneous fission of the Pu-240 contaminant in plutonium though.

To minimize this problem the presence of light elements (especially, beryllium, boron, fluorine, and lithium) must be kept low. This is not really a problem for U-233 used in implosion systems since the neutron background problem is smaller than that of plutonium. For gun-type bombs the required purity level for these elements is on the order of 1 part per million. Although achieving such purity is not a trivial task, it is certainly achievable with standard chemical purification techniques. The ability of the semiconductor industry to prepare silicon in bulk with a purity of better than one part per billion raises the possibility of virtually eliminating neutron emissions by sufficient purification.

U-233 has a spontaneous fission rate of 0.47 fissions/sec-kg. U-232 has a spontaneous fission rate of 720 fissions-sec/kg.

Despite the gamma and neutron emission drawbacks, U-233 is otherwise an excellent primary fissile material. It has a much smaller critical mass than U-235, and its nuclear characteristics are similar to plutonium. The U.S. conducted its first test of a U-233 bomb core in Teapot MET in 1957 and has conducted quite a number of bomb tests using this isotope, although the purpose of these tests is not clear. India is believed to have produced U-233 as part of its weapons research and development, and officially includes U-233 breeding as part of its nuclear power program.

Its specific activity (not counting U-232 contamination) is 9.636 milliCi/g, giving it an alpha activity (and radiotoxicity) about 15% of plutonium. A 1% U-232 content would raise this to 212 milliCi/g.

The key bits here being that we want to find natural Th that is low in Th-230 if possible but we also want to include a neutron moderating shell that cuts down the N to below the 500 KeV level. Ok, make sure that the “moderator” between the reactor 2 foot sphere and the ‘breeder’ sphere is effective. Maybe fill it with powdered charcoal briquettes if needed… Not exactly a major technical hurdle…

Also note the size of a core they talk about. 10 kg. So about 22 lbs is the production goal. Even at low production rates, the breeder blanket 3 to 4 feet in outer diameter and foot thick will hold a LOT more than that of Thorium, so number of runs will not be large (or better yet, a continuous processing fluid flow, but I digress…)

An Issue Of Timing

Note that the Wiki has the ‘concensus’ that U-233 will be too contaminated by U-232 to be usable.

Weapon material

The first detonation of a nuclear bomb that included U-233, on 15 April 1955.

(picture of folks looking at mushroom cloud deleted – E.M.Smith)

As a potential weapon material pure uranium-233 is more similar to plutonium-239 than uranium-235 in terms of source (bred vs natural), half-life and critical mass, though its critical mass is still about 50% larger than for plutonium-239. The main difference is the unavoidable co-presence of uranium-232 which can make uranium-233 very dangerous to work on and quite easy to detect.

While it is thus possible to use uranium-233 as the fissile material of a nuclear weapon, speculation aside, there is scant publicly available information on this isotope actually having been weaponized:

Note that they leave out the Indian test and the USA test… Wonder why… We’ve had U-233 bombs already made and tested.

The United States detonated an experimental device in the 1955 Operation Teapot “MET” test which used a plutonium/U-233 composite pit; its design was based on the plutonium/U-235 pit from the TX-7E, a prototype Mark 7 nuclear bomb design used in the 1951 Operation Buster-Jangle “Easy” test. Although not an outright fizzle, MET’s actual yield of 22 kilotons was sufficiently below the predicted 33 that the information gathered was of limited value.

Oh, there it is. Nice “damning by faint praise” guys… Not mentioned is the “swap” of material from U-235 to U-233 was a last minute thing and not approved, so the information of “limited value” was because they were looking for info on U-235…

The Soviet Union detonated its first hydrogen bomb the same year, the RDS-37, which contained a fissile core of U-235 and U-233.
In 1998, as part of its Pokhran-II tests, India detonated an experimental U-233 device of low-yield (0.2 kt) called Shakti V.

Yes, just ignore that as it is “low-yield”… nothing to see here, move along, move along…

Any ideas about WHY a twice proven to go boom “boom stuff” that is very similar to plutonium (though needing a bit more) would be poo-pood in such a way? I’m sure it’s not worth looking at… /sarc;>

In reality land, the MET test was a last minute swap of materials in a quest for information, not an optimization, and the India test was part of a series of “limit” exercises all done at once (as they figured they had one shot at getting past “monitoring”) that included a power reactor grade Pu bomb (supposedly not possible, yet it blows…) and was in some ways a test of ‘minimal to go boom’ so expected to be low-yield.

Bottom line is that the U-233 “boom stuff” works rather well.

The B Reactor and others at the Hanford Site optimized for the production of weapons-grade material have been used to manufacture U-233.[13][14][15][16]
U-232 impurity[edit]
Production of 233U (through the irradiation of thorium-232) invariably produces small amounts of uranium-232 as an impurity, because of parasitic (n,2n) reactions on uranium-233 itself, or on protactinium-233:
232Th (n,γ) 233Th (β−) 233Pa (β−) 233U (n,2n) 232U
232Th (n,γ) 233Th (β−) 233Pa (n,2n) 232Pa (β−) 232U
The decay chain of 232U quickly yields strong gamma radiation emitters:
232U (α, 68.9 years)
228Th (α, 1.9 year)
224Ra (α, 3.6 day, 0.24 MeV)
220Rn (α, 55 s, 0.54 MeV)
216Po (α, 0.15 s)
212Pb (β−, 10.64 h)
212Bi (α, 61 m, 0.78 MeV)
208Tl (β−, 3 m, 2.6 MeV)
208Pb (stable)
This makes manual handling in a glove box with only light shielding (as commonly done with plutonium) too hazardous, (except possibly in a short period immediately following chemical separation of the uranium from its decay products) and instead requiring complex remote manipulation for fuel fabrication.
The hazards are significant even at 5 parts per million. Implosion nuclear weapons require U-232 levels below 50 PPM (above which the U-233 is considered “low grade”; cf. “Standard weapon grade plutonium requires a Pu-240 content of no more than 6.5%.” which is 65000 PPM, and the analogous Pu-238 was produced in levels of 0.5% (5000 PPM) or less). Gun-type fission weapons additionally need low levels (1 ppm range) of light impurities, to keep the neutron generation low.

The Molten-Salt Reactor Experiment (MSRE) used U-233, bred in light water reactors such as Indian Point Energy Center, that was about 220 PPM U-232.

The point? Figuring out how to minimize the U-232 content of your U-233 matters. Rather a lot. So where does it come from and how to minimize it?

Alongside its abundance, one of thorium’s most attractive features is its apparent resistance to nuclear proliferation, compared with uranium. This is because thorium-232, the most commonly found type of thorium, cannot sustain nuclear fission itself. Instead, it has to be broken down through several stages of radioactive decay. This is achieved by bombarding it with neutrons, so that it eventually decays into uranium-233, which can undergo fission.

As a by-product, the process also produces the highly radiotoxic isotope uranium-232. Because of this, producing uranium-233 from thorium requires very careful handling, remote techniques and heavily-shielded containment chambers. That implies the use of facilities large enough to be monitored.

The paper suggests that this obstacle to developing uranium-233 from thorium could, in theory, be circumvented. The researchers point out that thorium’s decay is a four-stage process: isotopically pure thorium-232 breaks down into thorium-233. After 22 minutes, this decays into protactinium-233. And after 27 days, it is this substance which decays into uranium-233, capable of undergoing nuclear fission.

Ashley and colleagues note from previously existing literature that protactinium-233 can be chemically separated from irradiated thorium. Once this has happened, the protactinium will decay into pure uranium-233 on its own, with little radiotoxic by-product.

“The problem is that the neutron irradiation of thorium-232 could take place in a small facility,” Ashley said. “It could happen in a research reactor, of which there are about 500 worldwide, which may make it difficult to monitor.”

The researchers note that from an early small-scale experiment to separate protactinium-233, approximately 200g of thorium metal could produce 1g of protactinium-233 (and therefore the same amount of uranium-233) if exposed to neutrons at the levels typically found in power reactors for a month. This means that 1.6 tonnes of thorium metal would be needed to produce 8kg of uranium-233. They also point out that protactinium separation already happens, as part of other chemical processes.

At this point, what is needed is an awareness of the timing of isotope conversions. Th to Pa to U. What are the timings?

From above, we see that Th-232 goes through a decay path to U-233 that has a ‘long pause’ at the Pa-233 to U-233 transition.

Th-232 + n -> Th-233
Th-233 -> (22.2 min, beta) -> Pa-233
Pa-233 -> (27.0 day, beta) -> U-233

Notice that Pa-233 has a 27 day time for conversion to U-233? So you can suck out the Pa early, wait a week or three, and then separate the U-233 as a relatively clean product. (Yes, I’m leaving a detail here a bit opaque… but the astute student will see the timing issues.)

In short, it is very ‘doable’to make a U-233 “boom stuff” device via a thorium cycle and with only chemical processes. It takes a unit about the size of an 18 wheeler truck, and it does not need a lot of things that draw attention other than a few gallons of heavy water at the very start.

Frankly, this scares the hell out of me and I’d love to see a proof that this pathway is completely bogus.

Subscribe to feed

About E.M.Smith

A technical managerial sort interested in things from Stonehenge to computer science. My present "hot buttons' are the mythology of Climate Change and ancient metrology; but things change...
This entry was posted in Nukes, Science Bits and tagged , , , . Bookmark the permalink.

40 Responses to The DIY Garage Nuke Method – Or how to make a bomb without really trying…

  1. DocMartyn says:

    Before you make hasenpfeffer, you need to catch your rabbit.
    Now look at radiation cross sections;
    As you state, you have to separate out the Proactinium and allow it to undergo beta-decay.
    You basically need a steady state system where you withdraw the mixed, radioactive, salt and remove the Pa and put the Th back, constantly.
    Now you have a radioactive solution of mostly Thorium, Uranium and Proactinium nitrates, plus some real nasties released from etching your tank. How do you separate out the Pa and yet be able to recycle the Th?
    Thorium can only really be dissolved in HCl or nitric acid plus fluoride.
    You can’t have Cl in the presence of neutrons as you get radioactive, corrosive, sulphide.
    Dick around with Th nitrate and you have a mixed insoluble oxides.
    So to recycle the Th back in to you reactor, you have to get the Pa out, quickly, without changing the anions, as Th will come out of solution first.
    Obviously you can start out with a huge amount of Th, continuously add and withdraw small aliquots from your reactor, age for six months and then separate the U from the Th; which is what everyone did.
    The chemistry is a bitch.

  2. M Simon says:

    Iran had a Polywell Fusion program. They have either quit that or the program has gone dark. A Polywell of the right design can produce copious neutrons.

  3. M Simon says:

    I should add that I did an amateur study on how Polywell could aid proliferation some years back. It is still out there if you know where to look.

  4. M Simon says:

    I was trained in the Naval Nuke program (’66). We all knew about SL-1. This is the first I have heard about LOPO, HYPO, and SUPO that I recall. Negative space indeed.

  5. M Simon says:

    Moderator shells: about 2 inches thick of ordinary water would do. For better neutron economy a little thicker shell of D2O could be used. Such a shell is also a good reflector so you get a smaller reactor volume.

  6. M Simon says:

    complex remote manipulation for fuel fabrication.

    Except that robots are getting cheaper all the time. You also might want to look into NBS (NIST) experiments in producing machined spheres using interferometers to control movement. Without much trouble (relatively) you should be able to control things to within 1/4 wave length of light (125nm) if you have good enough temperature control. A look into the machining of IR telescope mirrors could also help.

    Controlling air temps to within 0.1 deg C is not difficult. And 0.01 deg C is doable with effort. You want to machine slowly with lots of air flow.

    If you know where to look and what to look for -as the Chief says – it is all out there.

  7. M Simon says:

    A 27 day decay half life says you wait – 27 day times 10 for 99.9% conversion. That is 1,000 ppm of your starting material left. Times 20 (540 days – about a year and six months) and you are down to 1 ppm. If you need extreme purity just wait. A two years and 3 months wait gets you to ppb.

  8. p.g.sharrow says:

    Think I’ll pass on building one of these. Too much work for something I have no need for. Although a nice self powered water heater might be nice to heat with or provide a hot tub bath! . 8-) pg

  9. E.M.Smith says:


    I was fascinated by the ORNL use of one for 150 kW of power and others being in the couple of kW range. It really would be possible to build a “house sized” nuclear heater and electricity plant. A 4 or 5 kW job would be just dandy, but I could get by on a 2 kW thermal / 1 kW electric constant supply and batteries.

    Oh, and I likely ought to mention that in addition to using the reactor to breed U-233, it would also be possible to breed Pu with it. The neutrons are there in any case. Plus, the government nicely provides tons of depleted Uranium for cheap (and that also means PURIFIED, with both the U-235 and U-234 removed). So one ought to get a very clean conversion of U-238 to Pu-239. Then it is just a fairly standard reprocessing step, but with far less of the other crap that is in spent nuclear fuel as this was irradiated just to get fuel, not to split atoms. (Yes, some will have split anyway… but less than in spent reactor high level crap).

    I’d go so far as to speculate that there is a simple trade off here. Chemical ease of processing (yes, it’s “a bitch”, but far easier than isotope purification) vs efficiency of conversion and higher radiation production. Short fast exposure to neutrons giving a low rad high purity product, but with more “turns” needed to get quantity. High exposure long time duration giving more radio-actives to deal with but a faster time to product and fewer “turns” needed. Somewhere in the middle ought to be the ‘sweet spot’ of max product for least total effort.

    Now if one has a few decades to run things, having a slow many turns production doesn’t really matter all that much. As long as it can stay hidden. (Though I suspect the gamma emissions even from a small operation would require it to be buried fairly deeply to avoid detection.)

    In any case, I find the “couple of foot” reactor size fascinating and the “few gallons of heavy water and you are good to go” boot up of interest too. Essentially, this puts building a nuclear power plant in the size range of even the smallest government on the planet with just a couple of folks with clue on staff. Even somewhere like Madagascar could afford and run a program at that scale. And get a load of electricity in the process. Breeding more fuel optional… but attractive, IMHO.

    Essentially it takes the Minimum Economic Scale down from the MegaWatt range into the KiloWatt range.

  10. DocMartyn says:

    (yes, it’s “a bitch”, but far easier than isotope purification)

    Actually I would rather purify Uranium isotopes via a cascade to about 75% and then do the final weapons step using laser isotope separation. It is really easy, cheap and can be done by monkeys. Alas, the monkeys in Iran are dumber than elsewhere and even with Khan’s stolen plans they still can’t do it right. A cascade running at steady state will consume less power than the lighting. That is the simple way to do chemistry.
    Seriously Chief, separating Th and Pa, so that you can quickly reuse the Th is none trivial.
    If you did the laser isotope separation on chloride as 35Cl absorbs a neutron to become 36Cl, then degrades by beta decay to 36S, which then decays back to 35Cl; eating your tank. Using chloride is a hell of a lot better than nitrate.

  11. E.M.Smith says:

    But but but… (he says arguing from ignorance…) Doc, folks have demonstrated chemical separation many times and it is asserted to be easier… Then again, TPTB have little interest in real relative issues being “outed” and do indulge in misdirection…

    I’ve not looked at laser isotope separation, so I’m clueless about it, just know “it exists”. Maybe I ought to give it a look… in my copious “spare” time… I do know that cascades are very large, take lots of power, and draw a lot of attention. A guy with a lab not so much.

    Quoting from the bottom of that last link in the above:

    “The problem is that the neutron irradiation of thorium-232 could take place in a small facility,” Ashley said. “It could happen in a research reactor, of which there are about 500 worldwide, which may make it difficult to monitor.”

    The researchers note that from an early small-scale experiment to separate protactinium-233, approximately 200g of thorium metal could produce 1g of protactinium-233 (and therefore the same amount of uranium-233) if exposed to neutrons at the levels typically found in power reactors for a month. This means that 1.6 tonnes of thorium metal would be needed to produce 8kg of uranium-233. They also point out that protactinium separation already happens, as part of other chemical processes.

    The first paragraph is clearly demonstrated above, IMHO.

    The second paragraph asserts that 8 kg is “enough” and that means 1.6 Ton of Th processed, in one month chunks in the breeder blanket. Don’t know the mass of a 3 inch thick shell at 3 foot inner diameter, but it is “not small”, so not a whole lot of batches needed. At a one month irradiation (or even two if outside the moderator the flux is lower than they are assuming) you just take that batch out and do the chemical separation and put in a new batch. (I’d use new batch each time and cycle the ‘bred’ stuff on to use in a running reactor as fuel, if at all possible). This means not a lot of “reactor isotopes” to deal with, just the breeder blanket isotopes. So ought not be a lot of container erosion from slow neutron absorption.

    Then there is only one line on the chemistry, and it doesn’t say it’s going to eat your lunch and takes a giant factory with remote manipulators. Then again, it doesn’t say it’s easy either…

    But is clear is doable, and has been done. Guess I need to do a bit more “dig” there…


    A separation of protactinium from neutron-irradiated thorium
    S.J. Lyle, A.D. Shendrikar

    A convenient-method, based on liquid-liquid extraction with N-benzoyl-N-phenylhydroxylamine in chloroform, is given for the separation of protactinium-233 from neutron-irradiated thorium.

    “Convenient” does not sound like “very very hard” to me…

    Since my “design” ( and I use the term VERY loosely ;-) has a separate reactor and breeder shell, we can run a different ‘mix’ in the breeder shell, even a MSR type fuel if desired, so this link per fluorides is potentially useful:

    Long, but interesting:

    It is, therefore, an object of the present invention to provide a method for the rapid and continuous removal of protactinium from a neutron-irradiated solution of thorium in a fused salt reactor fuel composition. A general object of this invention is to separate protactinium from a fused salt composition containing protactinium and thorium and/or uranium. A specific object of this invention is to increase the neutron economy in a fused salt reactor system designed to operate in conjunction with a thorium-232-uranium-233 breeding cycle. A further object of this invention is to separate micro quantities of protactinium from relatively large quantities of thorium and/or uranium. Another object is to provide a quantitative and selective method of separating protactinium from a neutron-irradiated halide mixture containing thorium and/ or uranium. A still further object of this invention is to provide a method of treating a fused halide reactor fuel containing thorium and/or uranium to selectively separate protactinium therefrom without essential- 1y changing the relative amount of ingredients comprising said fuel.

    With these and other objects in mind, the present invention, in its most useful aspect, consists of a method of separating protactinium from a neutron-irradiated fused fluoride reactor fuel composition comprising the steps of converting the said protactinium in said composition to an insoluble oxide product by contacting said composition with an inorganic metal oxide which is thermally stable at the temperature of said composition and which is thermodynamically less stable than the resulting protactinium oxide produced, and thereafter separating the resultant insoluble protactinium oxide product. By thermodynamically less stable is meant that, under the conditions of contact, the inorganic metal oxide precipitant will be fluorinated to form a soluble metal fluoride and the protactinium in solution will be selectively and quantatively converted to insoluble protactinium oxide product. Among the inorganic metal oxide precipitants which are operable to selectively convert protactinium in molten fluoride compositions to a separable insoluble oxide prod not are alkaii metal oxides such as lithium oxide, sodium oxide, rubidium oxide; alkaline metal earth oxides such as beryllia, and calcium oxide and refractory metal oxides such as zirconia, alumina, uranium oxides, and thorium oxide.

    While a relatively large number of metal oxides are useful from the point of view of selectively precipitating protactinium values from a fused halide solution, there are several other considerations which limit the practical choice of available precipitant metal oxides to a relatively small class when the process of this invention is up plied to a neutron-irradiated fused fluoride composition. A metal oxide which can cause selective precipitation of protactiniurn values cannot have an objectionably high neutron capture cross section, or it would defeat a major object of the invention which is to increase, or at least maintain, the neutron economy of the system to be treated. in another case, the added metal oxide may be satisfactory from a nuclear standpoint, but may be unsatisfactory because of its effect on the physical character of the system. For example, the addition of a selected metal oxide precipitant may efiectively precipitate protactinium, but may aiso adversely affect the melting point, density, viscosity or heat transfer properties of the system.

    It is, therefore, apparent that of the broad class of metal oxides which may be effective to selectively precipitate protactinium values only those metal oxides which have the least deleterious effect on the physical and nuclear characteristics of the protactinium-containing system can be used with maximum advantage. We have found that the oxides of the corresponding metal fluorides comprising the molten salt reactor fuel system to be treated can be used with maximum effect to selectively precipitate protactinium values and at the same time cause a least disruptive effect on the physical and nuclear properties of the treated system. Thus, for example, in treating a neutron-irradiated fused fluoride composition consisting of lithium fluoride, beryllium fluoride, and thorium fluoride to precipitate and separate protactinium therefrom, the most advantageous metal oxide precipitant to be used should be selected from the group lithium oxide, beryllium oxide, and thorium oxide. Similarly, in a neutronirradiated fused salt mixture consisting of lithium fluoride, beryllium fluoride, thorium fluoride and uranium fluoride the most advantageous protactinium precipitant metal oxide should be selected from the group lithium oxide, beryllium oxide, thorium oxide, and uranium oxide.

    The method of the present invention is extremely sensitive. We have found that fused fluoride compositions containing less than 10 parts protactinium per billion parts of total metal fluoride content may be treated to separate and recover substantially all of the total protactinium content.

    I would be interested in finding a fluid that could contain the salt at lower temperatures than having a molten salt, but that’s a bit of ‘finesses’ that might be a dead end. The key point of this bit being that it looks like folks have figured out some direct ways of getting Pa out of mixes.

    A different method (again using MSR salts… I’d hate to need to convert things to the fluorides first, so would look for other ways, but even that is not ‘beyond the pale’…)

    United States Patent O US. Cl. 23325 9 Claims ABSTRACT OF THE DISCLOSURE A method for reprocessing spent molten metallic fluoride salt mixtures containing uranium is provided whereby uranium and bred-in protactinium values are selectively extracted from the salt phase containing rare earth fission products into an immiscible metal solution by controlling the reduction potential between the metal and salt phases. Uranium values, which are extracted first, are oxidized and transferred from the metal solution into a barren salt mixture for ultimate recycle to the reactor prior to extracting the contaminants. The protactinium values are then extracted into the metal solution, oxidized, and transferred from the metal solution into a molten salt outside the reactor environment to await decay of the protactinium-233 to uranium-233.
    Previously a process was demonstrated for reprocessing a two-region, two-fluid molten salt mixture. In the fuel salt containing uranium and rare earth fission products but not thorium or protactinium the uranium values were removed first from the molten fluoride core salt by fluorination. Then the rare earths were separated from the lithium-beryllium fluoride mixture by extraction into a bismuth-lithium solution. This was proposed because it was thought in the art that insufiicient difference in reduction potentials for uranium and rare earth existed to permit the separation of uranium from the rare earths by this process. Furthermore, it was not thought possible to construct a single-fluid molten salt breeder reactor because it was not believed that there existed diiferences in the extraction potentials suflicient to separate uranium from protactinium and thorium and rare earth and protactinium from thorium and rare earths. It is therefore highly desirable and an object of this invention to provide an eflicient method for reprocessing spent molten salt fluids of protactinium and fission product values which obviates the separate fluorination removal operation for uranium and also allows the construction of a singlefluid molten salt breeder reactor.

    SUMMARY OF THE INVENTION This object was achieved by the discovery that an electromotive series for uranium, protactinium, and thorium and the rare earth fission products did exist which could provide suflicient difference in reduction potentials to permit the selective, reductive separation of uranium and protactinium values from the molten fluoride salt by an extraction process. One embodiment of the invention comprises contacting a molten metal fluoride salt mixture, such as LiFBeF ThF,UF (67.6820.012.0- 0.32, mole percent) useful as a single fluid, double region fuel, with molten bismuth and adding incrementally to the salt mixture quantities of a metal reductant which is less noble than protactinium whereby the uranium, and protactinium values are sequentially reduced to metal, transferred into the metal phase, and separately recovered by bubbling hydrogen fluoride through recovery molten fluoride salt mixtures, which are brought in contact with the metal phase, or by using other oxidation methods, thereby oxidizing the species to their respective fluorides and transferring them into the salt phase. The uranium values may be recycled to the reactor for further service and the protactinium values retained in a separate molten salt mixture awaiting their decay to uranium-233.

    It may thus be seen that by this method the removal of uranium in a separate, highly corrosive, time consuming, processing operation by fluorination prior to extracting the protactinium values from the salt mixture is obviated, Moreover, the method of the invention lends itself to high volume throughput consistent with minimum space availability within the reactor containment facilities and is readily adaptable to remote processing with a minimum of control.

    I’ve worked with hydrofluoric acid and know it’s a pain (literally… don’t get it on you…) but the methods are pretty well known, so even if you take a “nitrate to fluoride” conversion (that I’d try to avoid) it’s not really all that God Awful. is in German, and my German sucks, but I think it is saying that there is a preferential absorption method:

    Diese Einsicht ist seit langerem bekannt (7), und es wurden daher in der Vergangenheit
    schon mehrere Methoden zur Pa-Vorabtrennung vorgeschl^en:
    — selektive Extraktion mit langkettigen Alkoholen (8)
    — Mitfallung an Braunstein (9)
    — Adsorption an Silicagel oder Vycorglas (10), (27), (28).
    Nach Literaturuntersuchungen stellten sich die Adsorption an sog. Vycorglas und Desportion
    mit Oxalsaure als das erfolgversprechendste Verfahrensprinzip heraus. Die beiden erstgenannten
    Methoden ergaben unbrauchbare Resultate. Aber auch das dritte Verfahren
    konnte nach den Literaturdaten noch in keiner Weise befriedigen, vor allem fehlten fast alle
    fiir die technologische Auslegung eines Verfahrens erforderlichen physikalisch-chemischen
    Daten. Angesichts der Tatsache, daB die Wiederaufarbeitung fernbedient hinter einer starken
    Strahlenabschirmung stattzufinden hat, ergibt sich fiir die adsorptive Protaktiniumabtrennung
    ein erheblicher Vorteil dadurch, daB sich diese Verfahren im Kolonnenbetrieb mit
    einem relativ geringen technischen Aufwand kontinuierlich gestalten laBt. Das als geeignetes
    Adsorbens in Aussicht genommene Vycorglas zeichnet sich auBerdem durch seine hohe
    Strahlenbestandigkeit, reversible Adsorptions-Desorptionseigenschaft sowie seinen relativ
    niedrigen Preis aus. Es handelt sich um ein Spezialglas hoher Oberflachenentwicklung mit
    ausgezeichneten Sorptionseigenschaften gegeniiber Protaktinium.
    Die wichtigsten Eigenschaften des Vycorglases, Type 7930, sind in Tabelle 2 zusammengestellt:

    Click to access 3445605754797.pdf

    Looks like an English description of a similar thing, and specifically references nitrate, so that Uranyl Nitrate fuel is looking like it can be done without a fluoridation… :

    Two methods are being investigated for the separation and recovery of protactinium from short-decayea thorium fuel in nitric acid solutions. The Pa233, thorim, wid uranium may be coextracted from highly acidic feed solutions with sjO$ tributyl phosphate, or the protactinium may be preferentially adsorbed on pulverized unfired Vycor glass or silica gel. Major effort has been on the adsorption method.

    Adsorption experiments with tracer concentrations of Pa*’3 in nitric acid solutions showed distribution coefficient maxima for protactinium of about 1000, 325, and 175 from 6 to 10 M HNO3 for laboratory-prepared silica gel, un-fired Vycore and commercial-silica gel, respectively Unfired Vyeor, a commercial, leached borosilicate glass containing %$ Si0 and about 3% B203, was used for most of these preliminary studies. Fired Vycor glass adsorbed little or no protactinium.

    The adsorption coefflcient of protactinium by unfired Vycor glass from nitric acid solutions increases as the contact time increases or as the particle size of the glass decreases and is dependent; on the concentration of salt or nitric acfd in the solutian. The adsorbed pro-tactinium may be eluted with oxalic or tartaric acids. In column experiments, as much as 10 me;: of Pa231 per gram of Wired Vycor was adsorbed from 0.50 M Th, 11 M HN03, 0.1 M Al, 0.10 g/liter Pa2s1, ad cQ.1 M F. With a column loading of 2mg/g glass, more than 9% of the protactinium was adsorbed from a similar feed solution containing 0.05 glliter of ~3231~ % increasing the loading to about 5 mg pa/& glass, a total of about 9576 of the protactinium had been adsorbed. Greater thn 99% was eluted wi%h 0,s M oxalic acid, *$ of which was eluted at a concentration of 4.1 mg/rds F’a231. The protactinium concentration in the feed soJ-u”&ions were stable, showing no change other than the klO$ analytical sraria%%an after standing one month at room temperature in plastic containers. Although the optimum conditions for column operations have not been determined, decontamination factors of protactinium from thorium, uraniwn, ruthenium, zirconium-niobfum, and total rare earths of 6 x 103, le6 x 104, 4 x 103, 3, and 5.8 x 105, respectively, hve been obtained in tracer experiments.

    Batch countercurrent scouting experiments with tracer protactinium showed that about 90% of the: protactinium was extracted with the thorium and uranium from 5 M HH03-J M rU(N0 ) factor of: 70 from ruthenium and about 10^5 from rare earths. of the zirconium-niobium extracted also. A partitioning strip or a second solvent extraction cycle may be provided to separate protactinium from the uranium and thorium and to increase decontamination from fission products.

    And then…


    The Consolidated Xdison Thorium Reactor fuel is to be irradiated to an average of 18,000 Mwd/metric ton in a 2-year cycle. At the time of discharge, the fuel will contain about 66 kg of uranium, 1 kg of protactinium, and 18 kg of fission products per metric ton of thorium. The protactinium in the feed represents only about 1.5% of the potential fissionable material, but it contributes about half of the activity of the fuel solution. Removal of protactinium from the dissolved fie1 prior to extraction of the uranium and thorium not only provides a means of recovering the protactinium but also decreases difficulty in the solvent extraction cycle. Adsorption of protactinium on unfired Vycor glass or silica gel is being investigated for this application.

    An alternative to this is the coextraction of uranium, thorium, and protactinium from dilute aqueous solutions with 36 tributyl phosphate (TBP). Laboratory experiments with simulated feed solutions have indicated satisfactory separation of products from the high-cross-section fission products by both these methods, but neither method has been sufficiently investigated that it may be recommended for processing actual reactor fuels. This report will be concerned primarily with the adsorption of protactinium on unfired, pulverized Vycor. In addition, some preliminary silica gel adsorption and solvent extraction data will be presented,

    I’ve noticed some odd artifacts from the cut / paste from the PDF but don’t have the time to fix them all, please hit the link / original for a clean read.

    The point of all this is just that the processes described, while sometimes a bit exotic (adsorption on glass? Really?) also doesn’t look that horrid. I’m sure there’s a lot more and I’m sure the devil is in the details, but I don’t see it stopping the small country or megalomaniac billionaire from doing it.

    It looks like it could even be as easy has having an aqueous nitrate solution flow over powdered unfired Vycor glass or silica gel. (or as hard as a fluoridation and all).

    So am I reading all this wrong?

  12. Larry Ledwick says:

    One other issue is a possible source for uranium other than known mining deposits.
    A country that wanted to acquire the raw uranium ore quietly outside normal commercial channels could use extraction from sea water (an already proven technology) even if it is more expensive it could be done very very quietly on ocean going ships or piggy backed on any facility that has by its nature ocean water access..

    Although centrifuge systems are the most well known, there are other avenues which could be pursued, and a desperate governments might stumble on a novel and highly efficient method which no one has give much thought to.

    For example the gas chromatograph depends on the fact that lighter isotopes defuse more quickly than heavier isotopes. This is used in the gas diffusion method forcing the material through layers of membranes where the lighter isotope diffuses faster than the heavier isotope. This was the system used in our first nuclear program while other methods were researched. It would side step a limit on centrifuge technology.

    The centrifuge is well known and understood but even there it can be improved with careful design and engineering. Exactly where certain countries are in that scale is likely unknown. It would not surprise me to find certain governments carrying on multiple research efforts to slowly but surely advance their designs outside the inspected facilities.

    It is only reasonable to assume that sooner or later some otherwise second class player will stumble on a major breakthrough due simply to their willingness to try unknown or poorly researched options that might be easier to hide from inspection. Things like combining previously known techniques in novel ways might allow substantial jumps in output or novel paths to a type of “boom stuff” which is perhaps a mix of several isotopes rather than any one pure isotope.

    Under estimating the opponent is a good way to find out you are a day late and a dollar short at a most inconvenient time and place.

  13. gallopingcamel says:

    Getting hold of a few hundred grams of Pu239 or U233 is not an easy task. Much more difficult than getting a few dozen Krytrons (as I did in the early 1970s).

    There are easier ways to build nuclear reactors. Charles Bowman built one on his farm in Virginia using black pine as the neutron mirror.

    Ed Bilpuch was one of my golf buddies at Duke university and he was a great leg puller. When he told me about the wooden nuclear reactor I laughed out loud. The next day Ed showed me the reactor which was being tested at TUNL (Triangle Universities Nuclear Laboratory). It was about eight feet tall and eight feet in diameter.

    This was an ADR (Accelerator Driven Reactor) that can be turned on and off like a light.

  14. E.M.Smith says:


    Yes, it’s very easy to make a reactor. Any pile of U-natural with moderator that is big enough will eventually “go”. The goal I had here, though, was to find a way that was very small, would work with natural Uranium for startup, and didn’t need materials likely to draw attention. Then, in addition, have a very safe operation profile.

    Because of how hard it is to get a chunk of U-235, Pu-239, or U-233; I was deliberately looking at reactors that can start on plain U-natural mix and then have a cycle that lets you breed up the boom-stuff isotopes and move to light water moderator (or similar) on the mix. Oh, and note that U-233 can go into either implosion type or gun type devices, so no need for krytrons.

    The aqueous homogeneous reactor has all that. If it goes to high power, the moderator boils out and it shuts down. On an approach to too high a power, the moderator expands and it slows down (eventually stopping). In short, it’s hard to screw it up for an amateur working in a primitive environment.

    I’d first considered the original graphite design, but given some of the core reactivity issues and tendency to burn (see Chernobyl) and the fairly large size needed, along with operational issues like fuel rod and control rod fabrication and operation, it was not a great choice.

    Compare the A.H. reactor with “fuel in a bucket” and moderator mixed in. Control rods optional (use a faucet on the side if desired…). Very forgiving operational profile. Core of a couple of feet size (though I might want to make it a right circular cylinder with H = D for ease of construction), and low need for anything “special” and that being a small can of heavy water, about what is used in biological experiments (on things like how much D2O slows down metabolism in plants), and it’s all a nice package. ( At about 40% to 50% D2O your enzymes start to not work quite right and some systems slow too much; over that you die. At least, if you are a mouse…)

    Yes, having an accelerator driven reactor makes it easier to control the neutrons, but folks tend to notice when you build a nuclear accelerator in your back yard.

    Like the Pine Reactor story. Web search only turned up some other folks making comments, not a write up. It would be interesting to find a site memorializing it with photos ;-) I DO like the idea of a “furniture wood reactor”… It appeals to my sense of “artistic strain” between elements.

    Picture a nice large polished furniture look ‘Chinese puzzle’ cube with some pipes coming out of it (for heat output / cooling)… Kind of a Steam Punk Reactor ;-) Now THAT would be way cool!

    But I would expect the same kinds of flammability and reactivity issues as with graphite reactors. Perhaps made more “interesting” by humidity / wood moisture change issues… Still, it’s one of those images that, once seen in the mind’s eye, will never go away. Thanks for that! ;-)

    @Larry Ledwick:

    I have a posting in process on sea water extraction. There have been some advances. Coming soon…

    I’ve sometimes wondered why one could not have diffusion membranes between centrifuge steps as a “kicker” since you already have the pumps and all running, but figured the Engineers had their reasons (maybe materials mis-match, or the diffusion step just makes the waste stream from that step not quite suited for recycle down one, and then you need some other process… or maybe just the KISS principle…). I do agree that “new ideas” tend to break “old regulations” before they are noticed.

    One of the things that first put me on this path was noticing that every time Th was mentioned, there was either a stunning silence, or folks on the ‘inside’ would poo-poo the idea; then looking into it, I’d find old information about it working “way back when” then disappearing… that “negative space” just begged for answering “why?…” Then the Indian U-233 bomb made that kind of clear. I’m pretty sure they just shoved some Th through their CANDU like reactor as a breeder and the processed it “right quick” before it got too hot. Proof of concept and all.

    That, then, points at “how to make a CANDU lite”, that leads to the Aqueous Homogeneous history… and add a breeder blanket with moderator…

    Frankly, I’m pretty sure that a similar approach was taken by some of the Indian and Pakistani work, but using a research reactor to slowly breed ‘boom stuff’ in a large pool type environment. Little at a time and you are there eventually. Part of what makes me worried about the “Deal” now on the table that leaves the Iranians with research reactors… But I digress.

  15. Larry Ledwick says:

    Your comment about negative information rings true to me.

    I recall a small item in I believe Popular Science or one of the similar magazines in the late 1980’s or early 1990’s talking about a new development in explosives initiation. It was just one of those little side bar items they use as fillers. It was a very very simple design improvement, something you could probably make in a basement shop in just a few days of tinkering. it used technology which was very mature for home experimenters in electronics (circuit board etching etc. ) and a slightly different design for the part that goes poof when it receives an electrical pulse. One of its advertised features was much faster initiation of the explosives than the typical bridge wire design, and as I recall better predictability in the time delay between the electrical pulse and initiation. The one key modification was a very simple substitution for the normal bridge wire.To date some 30+ years later I have never seen the slightest reference to this design any where else it just dropped into a black hole.

    If it performed as described, It would have been highly suitable for applications that need very quick initiation and high precision in timing of initiation (wink wink)

    One of the lessons you quickly learn when working in high security environments is to pay attention to what is Not said and How it is not said. Often that is where the information is. People have a natural tendency to try to talk around things that they should not talk about explicitly, and sometimes they are quite clumsy about how they do it.

  16. M Simon says:

    Yes, having an accelerator driven reactor makes it easier to control the neutrons, but folks tend to notice when you build a nuclear accelerator in your back yard.

    If you build a Farnsworth type accelerator hardly any one notices. They are not very efficient though. Now if you went to a Polywell (an extrapolation of a Farnsworth machine) you could do better. But things start to get large. You are now beyond basement and into garage or barn territory.

  17. punmaster52 says:

    I think you guys are overlooking the obvious here. Hillary Clinton has all kinds of contacts in Russia, and would probably get whatever materials you wanted for a substantial fee. Sorry, Chiefio, it’s the process, right, not the product?

    I realize all of you are technically oriented and consider this sort of thing just for the intellectual challenge, but aren’t biological, ahem, items, a good deal less complicated these days?

    ( goes to kitchen, gets cinnamon roll, makes resolution not to be further drag on the party atmosphere )

    Significant Personal Death Probabilities sounds like a good name for something. Too long for a band; maybe the first tour for Nuclear Processors, or Uranium Handlers?

  18. DocMartyn says:

    Chiefio, if you have a radiated blanket of Thorium metal getting the Pa out is easier than starting with the nitrate. The fluoride salt reactor has huge advantages in this area as you can do the oxide precipitation, which floats, and just circulate the U/Th.
    However, this stuff is very toxic and very radioactive. Every ‘wash’ in your purification is going to make a lot of low/medium waste water.
    The big powers have used uranium separation and plutonium separation for a reason.

  19. M Simon says:

    And then there is this:
    Scale it down and use it as a neutron spalling accelerator and you are in business. The question is – can everything be scaled down? GEVs are not required. BEVs will do nicely. And what is the efficiency of the whole works.

  20. M Simon says:

    punmaster52 says:
    27 April 2015 at 4:03 pm

    Bombs are local, biologicals are not. And then there is the problem of third world medicine vs first world.

  21. E.M.Smith says:


    IMHO it’s all about the neutrons. I’ve just assumed that electronic gizmos can’t give enough total neutrons at the same rate in the same size. I could easily be wrong. I’d thought of that path, but figured it would be a very very slow one due to low numbers of neutrons.


    Didn’t I say up there somewhere that the smart shopper would just take a suitcase of cash to N. Korea or Pakistan and walk away with a working device? I’m sure I covered that… ;-)

    Then the Russians, well, that’s just a matter of money and a “concession” or two in some nice big city with lots of “pigeons”…

    And absolutely, a nuke is likely THE hardest WMD to make / get. Bio is much easier and can be done in the kitchen. (For starters, there was a case of Anthrax that took out a dozen cows or so about 30 miles from me a decade back. The stuff lives in the soil all over. “Wool Sorters Disease” was the old name. So just find somewhere with lots of sheep and look for one with black scabby spots. Take sample. Grow. Now you need a freeze dryer and grinder (and antibiotics for WHEN you inhale some of it….) and you are “good to go”.

    Plague is endemic to the rodents of California. Go to Sierra Nevada, trap squirrels that look rough. Test until satisfied. Raise a large batch with “culture” ….

    There are more complicated approaches, but you get the idea. BTW, for a long time you could just call up bio supply houses and order refined cultures of things like U. Pestis and Anthrax. Lately they have required “papers” for you to do that. I’m sure it’s all secure now… /sarc;>

    Not to mention I can make chlorine gas with bleach and ammonia from the grocery store. (And even more interesting things with about the same level of effort).

    Cinnamon toast and coffee? Sounds like a great idea… I think I need to cheer up out of this maudlin topic..


    As I am a complete amateur in this whole area, I’d already disclaimed above (somewhat too much, I’d even speculate…). With that reminded, yes, having just Th in the blanket would be ‘easy’ too, relatively speaking, I suppose. But having it already in solution or slurry has processing advantages (and especially speed of processing advantages). MSR certainly IS better in this regard for a sophisticated operator, but the thesis here was more “backyard hack” where I’m suggesting (speculating?) that the temperatures and handling of molten salts are problematic for a backyard Joe, while liquid aqueous chemistry is likely familiar.

    I’d also point out the nitrate solution over powdered glass separation above seems pretty straight forward to me. (Then again, having never tried it nor had to work with radioactive liquids in bulk, I can see potential “issues”…)

    Hmmm…. The waste water tail… Hadn’t considered that… Need some kind of scrubber filter that doesn’t irradiate you or heat up…

    My speculation was that the “big powers” have chosen that path at least in part because it is hard and because it dissuades other from doing it. I think there is evidence of this in the pressure put on Canada to dump the CANDU early on and the avoidance of mentioning U-233 as “boom stuff” after the early days. (That India then used exactly those kinds of things to make a bomb, which they tested successfully, lends a further push in that semi-paranoia-direction…)

    Not saying I’m right, just saying I’m suspicious and I think it is worth picking at…

    @Larry Ledwick

    When I was about 18 there was an article in Pop Science about a Dentist who made an airplane using plastics and composites and how it was going to make personal airplanes much cheaper. Only problem was that only the engine showed up on RADAR so it was a bit of a traffic hazard. Then Nothing… About 20 or 25 years alter, digging into the history of stealth, found a reference to some of the early efforts coming from a Dentist… So I suspect he was “recruited” by a Three Letter Agency (perhaps against his will) and told, no, no royalties from private plane sales, but here is a nice retirement plan…

    I’d been on my way to MY Dentist while reading it, and very interested in planes, so the story stuck with me. For decades after I kept looking for more stories on plastic planes with low radar cross section. That was one of the first “deafening silences” that caught my attention…

    Then going to a school where most of the Lab Rats Kids (from LLNL) went as it was close, and secure, and had a decent number of Clarence Clearance types already in place; and talking with said kids, and visiting their families; well, you get a good sense of that “not saying while discussing” process. (one girl only knew about her Dad’s work that “He has a PhD Physics, works at LLNL, and has both a winter and summer suitcase. He will come home, grab one, and be gone for a few weeks. The one he grabs may or may not match the current season here.”)

    So in one dorm conversation I learned just how small a nuclear warhead for an artillery shell can be (smaller than was acknowledged publicly). In another, some ideas on novel designs. Etc.

    Then one day when a “30 something”, I was talking with my cousin ( PhD. Ops Research IIRC) who was a former Army Nuke and had “carried the keys” for Europe for a while… He asked “how small do you think a nuke can be?” (This in the “Russian Suitcase Nukes missing” rumor days.). I proceeded to describe a briefcase sized device, with Be tamper and annular confinement… His response? “HOW do you know that!?” …. Notice that in “confinement” designs folks ALWAYS talk about the difficulty of making a spherical implosion? NEVER about annular cylinder radial implosion confinement with inertial / explosive ‘end plates’? Now look back at some of the very old nukes tested on atolls. Notice the timing wires entering the RIM of a not quite spherical device…. Makes you go “Hmmmm”…

    (not sure if his reaction was to the confinement or to the mention of Be tamper, though…)

    No idea if there actually has been an annular confinement nuke ever made, and web searches don’t show up much. But you would think if it didn’t work folks would say so somewhere… Hmmm…

    Personally, I’d try to set it up with a “pit” of trigger at each end of the assembled cylindrical core (Be thin coat over neutron emitter metal that makes neutrons when whacked – detail omitted but in web descriptions of Russian devices) and try to set up a neutron ‘lasing’ action down the length of the ‘assembled’ cylinder. Alternatively, a long thin trigger down the whole length and a ring of imploding SNM that self triggers on assembly…

    Personally, I don’t see this as a ‘big leap’ since gun type U bombs use a cylindrical assembly but via a plug fired into a ring from the end with a back stop. I’m just changing the axis of assembly to ‘edge in’ and doing it explosively; optionally with a very large flat “barrel analog” in the Be end plates (held in place with explosives during “assembly”).

    At any rate, I’m rambling. It has just been a sort of mild interest for about 45 years now and I’ve gotten good at seeing the negative space scattered all over the topic. Now, with everyone and his cousin having or able to get a nuke, avoiding the topic seems a bit daft.

  22. E.M.Smith says:

    Actually, thinking about it, the cheapest, easiest, most effective WMD would be to just get one group of people all worked up an mad at another group of people and stand back. Kind of like Sunni vs Shia or Catholic vs Protestants a few hundred years ago, or even Muslim Radicals vs {everybody else} today. Or even like “Watermelon Greens” vs “Deniers” ;-)

    So many times in history a propaganda campaign has resulted in mass casualties in the millions. From Stalin to Pol Pot to W.W.I and more. It simply MUST be the most effective WMD around.

  23. Larry Ledwick says:

    I think the important lesson of history is there has never been a useful weapon system which did not eventually “leak” outside of the confines of a cloak of secrecy. Granted it took a long time for gunpowder to become well know chemistry, but they did not have an interweb back then either.

    When you have Princeton undergraduate students working out a sufficiently viable design to have the powers that be classify his work you know any intelligence agency worth their pay check has plenty of info to work with.

    I have an interesting book I picked up years ago on a discount book table for 99 cents
    It was titled “The Islamic Bomb” copyright 1981. It unfolds the entire story of how the Islamic world teamed up to get the Pakistani bomb built. It clearly shows that with enough money and patience any country on earth could have one if the wanted one. Heck South Africa built 6 before they decided they were more trouble than they were worth. (note enrichment method they used – limiting centrifuge numbers does not block enrichment!!)

    In 1977 the student published a design for a 125 pound Plutonium device which could be built for $2000 if you already had the SNM and only needed the structural hardware to make it work.

    As far as size goes the U. S. Military did field an atomic artillery shell for the 8 inch gun (203 mm)

  24. punmaster52 says:

    Well, yeah, you did say something about the easiest way being a suitcase full of cash, but that was waaay up in the beginning and two paragraphs late I forgot it. Short term, short term, short term something loss.

    After your comments on these of producing biologicals I wish I hadn’t mentioned it. Didn’t know it was that easy. The chlorine gas I did once before I knew how it worked.

    Cinnamon rolls. Toast isn’t food compared to the wifes’s rolls.

  25. punmaster52 says:

    That was supposed to be: on the ease of producing biologicals

    And wife’s, not wifes’s

    Where have my spelling skills gone this afternoon?

  26. J Martin says:

    OK, so now I’m confused. I thought one of the advantages of using Thorium as a nuclear fuel was that it reduced proliferation as it would take an awful lot of reactors to produce a bomb. But scanning through the above suggests otherwise ?

  27. E.M.Smith says:


    Might I ask that your wife talk to my wife about wifely duties? Somehow I ended up the one cooking all the time. While I’d love to have the wife make cinnamon [anything} it’s more like I make what I can. Clearly you have a better handle on this, so I beg, please…

    No worries, at least it is not wifes….

    @J. Martin:

    I really do recommend “The Curve Of Binding Energy”… and remember that deception comes with this particular turf… so TH in a MSR run for long term power production is ‘reduced proliferation’ but my whole thesis here is that “there is another path” that is hidden by distraction…

    Thus your justified “say what?” reaction…

  28. punmaster52 says:

    @ E. M. Smith:

    I doubt I can help. My wife does it to keep me from whining. She can’t stand that. :-)

  29. gallopingcamel says:

    “Actually, thinking about it, the cheapest, easiest, most effective WMD”

    The cheapest WMD I know of is currently being used to remove wrinkles from the faces of aging ladies thus proving that technology can be used for good or evil.

    My chemistry tutor was Dr. Saunders a small, dapper professor (he lectured in a three piece suit). He spent WWII weaponizing Botulinus toxin. One fifty gallon barrel is sufficient to kill the entire human race.

    The LD50 dose (50% of affected indivuiduals will die) is ~2 ng/kg. For a hefty chap like me the LD50 dose would be 200 ng.

    A fifty gallon barrel would contain about 200 kg of toxin or 1000 billion LD50 doses. Given that the human population is around 7.2 billion one barrell would be sufficient assuming an efficient delivery method.

    Saunders tested the toxin on himself (it was war time and OSHA did not exist).

  30. p.g.sharrow says:

    Actually Botulism toxin was used to treat my lady’s migraines and it seemed to work well to stop re-occurrence , so the treatment was stopped, Medicare stopped supporting these kind of treatments in favor of those that don’t work. pg

  31. M Simon says:

    p.g.sharrow says:
    28 April 2015 at 5:46 am

    Medicare stopped supporting these kind of treatments in favor of those that don’t work. pg

    All of US medicine is like that. The endocannabinoid system – fully exploited – could save between $1/2 trillion and $1 1/2 trillion a year. Try just three diseases – cancer $200 bn a year. – Alzheimer’s $200 bn a year – 2/3rs of Type II diabetes $200 bn a year. So my estimate is not unreasonable even if the treatments are only 50% effective. .

    The biggest supporter of “Drug Free America” is Big Pharma. Medicine in America is as corrupt as hell. Here is a short bit on that corruption.

  32. gallopingcamel says:

    My apologies to Chiefio fraorad “Off Topic” but I oul like to support what M Simon says. Here is something I wrote many years ago concerning the Medical Mafia:

    “In the fall of 1970 I visited the USA to exhibit electro-optic equipment at a trade show held in the New York Coliseum. On this, my first trip to the USA, I experienced more than just culture shock owing to arriving just in time to experience a hurricane; the Attica prison riots; plumes of smoke rising every day as apartment buildings were torched; dreadful roads; Carnegie Hall and other famous buildings falling apart. American gangster movies were quite popular in the United Kingdom back then but that did not prepare me for the reality of organized crime.

    Two large crates containing the equipment and literature needed for the show were sent over 3,000 miles by air from the United Kingdom to New York. My first task on arrival was to arrange transport for the crates from JFK to the New York Coliseum, a distance of 35 miles. The first trucking company quoted a price that was close to what I had paid for air freight across the Atlantic. The cost was so outrageous that I spent an entire day trying to get a reasonable quote from a dozen trucking companies until someone took pity on me and explained that there was only one price and it was set by La Cosa Nostra (LCN).
    In 1982 I emigrated to the USA to work in Manhattan. The freight rates out of JFK were still unreasonably high, a clear indication that the LCN had managed to retain control throughout the 12 years since my first visit. How was such a thing possible when it was common knowledge to everyone who did business via the New York airports? Over time I was to learn that the freight out of JFK was just a minor part of LCN operations involving 250 trucking companies supported by labor unions that controlled garbage collection, concrete, the garment business and much more.

    Although the services controlled by the LCN appeared to be run by independent companies there was no effective competition so from a customer’s standpoint it was like dealing with a monopoly. Consumers could complain though nothing would be done as long as the LCN was intimidating competitors while greasing its connections in government and law enforcement. The situation could have continued indefinitely if the FBI had not decided to mount a series of high profile operations targeting the Luchese crime family in the 1980s.

    While organized crime is still flourishing it is under increasing pressure from a variety of law enforcement agencies that use the publicity from prosecutions of major crime figures very effectively. Sadly, a new Mafia has risen up that poses a much greater threat to the people of this country than the old one ever could owing to its immense scale and the fact that it is not regarded as a criminal conspiracy. Nevertheless it involves the corruption of government at all levels and the exploitation of the general public, especially the most vulnerable of our citizens.

    If goods and services cost far more than is reasonable it is usually a sign that corruption is present. Corruption costs money and the cost is passed on to consumers. According to the National Coalition on Health Care medical expenditures in the USA amounted to $1.7 trillion in 2003 rising to an estimated $1.8 trillion in 2004. Our expenditures per capita are typically double those of our major trading competitors and comparisons are far less favorable with low cost providers of health care such as many Latin American countries. The rate of growth is even more alarming. In 1950 medical expenditures accounted for 5.2% of the GDP rising to 9.4% in 1975. By the year 2000 expenditures had risen to 15.4% of the GDP or roughly four times what is spent on national defense. In 2007, 62.1% of filers for bankruptcies claimed high medical expenses.[4] If the country is not capable of radically reforming the health sector the spending could well exceed 20% of GDP by 2025.

    To reduce this to things that matter to individual consumers I made comparisons of medical procedures for an uninsured person in Sarasota, Florida with comparable procedures in Mexico, Costa Rica and Colombia. Knee replacement surgery costing $46,000 in the USA can be had for under $10,000 in Costa Rica. A cranial MRI scan costing $700 in Sarasota not only cost much less in Bogota ($102) but involved much less hassle than one gets in the USA.
    How did things get so bad? According to several experts we are suffering from a perfect storm caused by the alignment of many powerful forces capable of manipulating the health care market to benefit themselves.

    The American Medical Association ensures high salaries for doctors by limiting the number of people trained in this country and placing barriers for foreign doctors who wish to practice here.
    The drug companies prefer developing products that will make money (e.g. Viagra) to working on ones that reduce human suffering.
    The Federal Drug Administration is a wholly owned subsidiary of the drug industry, working hard to prevent people from obtaining cheaper drugs.
    Health Management Organizations exist to make money while dragging down the quality of service experienced by the general public.
    Trial lawyers drive up the costs of malpractice litigation, getting rich while smugly pretending to be standing up for the victims of the medical colossus.
    The legislators produce laws that drive costs up while achieving little of value for anyone other than the lawyers and HMOs (remember HIPAA?).

    What can be done? The present system appears to be beyond hope of reform. The conspiracy is so huge and its members so powerful that there is no hope of it being declared a criminal enterprise. The medical industry will continue to increase its share of our GDP, pauperizing pensioners and destroying once mighty companies such as General Motors. At some point there will be a revolt; let it be soon.

  33. E.M.Smith says:

    @GallopingCamel & M. Simon:

    No argument from me!

    I worked in the medical field (as a clerk…) during some of that transition to TheDarkSide. Watched the “Old School” family and country Doctors and Nurses get a bit wide eyed and indignant at some of it… then say that retirement was not too far away so they would just put up with it.

    One clear example? The Pharmacopia. This is the official approved list of drugs. Go outside of it, you get trouble. (Or in HMOS get walked out). What fairly recently left it? GRAS by the ton. When young, there was effective and cheap tincture of Iodine, Merthiolate, Gentian Violet, Mercurochrome, and more. All dirt cheap and effective. Had a fair quantity spread on ring worms, athletes foot, cuts, etc. It was, for a while, Generally Regarded As Safe and kept in.

    Few years back, the FDA decided that GRAS had to be tested, or was out. Drug companies don’t do trials on generics that are dirt cheap… so they all are now, nowhere to be seen in the drug store (though, I’ve heard, you can get it underground from India…)

    Now don’t get me wrong, Lotrimin and others newer athletes fungus cures are much more effective. But do I really need a triple antibiotic cream at several dollars for a small tube to put on a scratch?

    Another casualty was various coal tar preparations. Good for scalp issues of many kinds (turns out creosote kills off all sorts of bugs, and inhaled smoke was used to treat and sometimes cure TB before antibiotics… remember that when the antibiotic resistant strain becomes dominant). Lucky thing, you can still get “pine tar soap”, just can’t say that it does anything but clean…

    In an even earlier turn, the various working botanicals were removed. I find it sad that today there is a medical / drug company jihad against herbal treatments when their own pharmacopia had them in it… but then were removed when new drugs came along.


    Popular passages

    Page 126 – Macerate the aconite root for forty-eight hours, in fifteen ounces of the spirit, in a close vessel, agitating occasionally ; then transfer to a percolator, and, when the fluid ceases to pass, pour into the percolator the remaining five ounces of the spirit. As soon as the percolation is completed, subject the contents of the percolator to pressure, filter the products, mix the liquids, and add sufficient rectified spirit to make up one pint.‎
    Appears in 38 books from 1859-2003

    Page 338 – Soap-bark for forty-eight hours, with fifteen ounces of the Spirit, in a close vessel, agitating occasionally; then transfer to a percolator, and when the fluid ceases to pass, pour into the percolator the remaining five ounces of the Spirit.‎
    Appears in 44 books from 1859-2003

    I have an e-copy in my bugout set, that’s presently in the back of the closet somewhere ;-)

    Oh, heck, I just started another PDF download…

    FWIW, in my seed archive I have a small section for “medicinal” seeds. For a while, one of the local shops had a small selection. Things like coneflower and chamomile and such. I have some poppy seeds from an organic food store (it is reputed that the food poppy can be useful…) but haven’t tried a germination yet. Many commercial kinds are sterilized, but I’m hoping the “organic” ones slipped through ;-) But I’ve been unwilling to try a ‘test grow’ for the obvious reasons.

    Part of my interest in ‘self cures’ and my stories about them comes directly from that observation, that perfectly good and working things were being ignored. So a tiny bit of my time goes into finding them and trying them ( IF I have an appropriate need ) and then saying what worked and didn’t.

  34. Larry Ledwick says:

    You cans still get those GRAS preparations online via
    etc. and in the local Walmart (Humco brand) They sell to an older crowd that still remembers these preparations

  35. Larry Ledwick says:

    By they way since we are on this tangent, a couple really good medical info sources

    410 page pdf by (appendix G has a full list of supplies for the dispensary and life boat kits etc.)
    In addition it has basic instructions for treatment of almost any ailment that is treatable in the field.

    Click to access ships.pdf

    Basic kit contents Mayo Clinic

    A doctors recommendation for an emergency kit for flight emergencies (and perhaps hard landings in strange places)

    Emergency War Surgery Paperback – January 1, 1992
    Book published by NATO for expedient field surgery and combat medicine
    (over the top for most situations but an interesting read)

    And :
    Medicine for Mountaineering & Other Wilderness Activities Paperback – February, 1993
    Backpack sized volume covering the types of emergency medicine encountered in mountaineering and similar back country adventures.

  36. E.M.Smith says:

    @Larry Ledwick:

    I think I have that Mountaineering book on the shelf… somewhere… or something very like it.

    The others look interesting too.

    BTW, not real worried about the ‘tangent’… This is an “outside the box” DIY thread after all. DIY nukes, DIY brain surgery, hey it all fits in the “what could possibly go wrong?” bucket ;-)

    Oh, and found the pharmacopia in a non-Google-watermarked-search-blocked etc. format. Your choice of text, PDF, ebook…

    British pharmacopoeia

    published under the direction of the General Council of Medical Education and Registration of the United Kingdom

    Published 1867 by Printed for the Medical Council by Spottiswoode in London .
    Written in English.
    Pains have been taken to make the descriptions of all the substances referred to in the work sufficiently comprehensive and minute to afford a clear indication of what the medicines of the Pharmacopoeia are intended to be, and to enable those who are engaged in their administration to determine the identity and test the purity of such as are met with in commerce. In the descriptions of natural products reference is made to their sources. When they belong to the animal or vegetable kingdoms, the scientific names of the animals or plants yielding them, if known, are given, in addition to the names under which they are used in medicine ; and reference is generally made, in the case of plants, to the best authorities for the scientific descriptions of them, and to works in which correct figures may be found. Mineral substances are described with reference to their chemical characters and composition ; and generally, in the descriptions of products, whether natural or manufactured, the distinguishing characters and tests are included, where such can be referred to with advantage.

    Yup, cut paste and highlight (bold) all working… and gotta love it… “animals or plants yielding them”… wonder how many MDs today know how to go to the botany department and womp up some meds… ;-)

  37. p.g.sharrow says:

    Coca plants, Afgan poppy seed, cannabis seed yield lovely house plants or ones that can be grown outside, if you need to. The seeds will remain viable for many years if kept cool and dry. pg

  38. Serioso says:

    I haven’t seen any mention of a dirty bomb. A nuclear explosion isn’t needed if the aim is to create panic.

  39. E.M.Smith says:


    Consider it mentioned. I mostly left it out as it is just so easy to do. Get medical equipment being recycled in Mexico (happened accidentally at least once that was found and fixed as the rebar in a building was, er, doing things…). Add to any convenient explosive package. Done.


    At a prior time I, um, er, ‘experimented’ with some of that. A nice little 1 shelf system with a 4 foot long ‘garage tube fluorescent’ fixture on the top, then a couple of layers of cinder blocks. A half dozen small plants as ‘starts’ in small pots on the very top level. Below that the larger pots. Nothing more than about 3 foot tall. More than enough for anything I was interested in and I ended up accumulating a quart or so of “excess” in just a few months.

    Things I learned:

    1) The Acetone Hashoil Extraction. Read my CRC (“the Rubber Book”) and found the magic bit was soluble in acetone. Visit to hardware store paint department later… One pint jar with dry green macerated bits, saturated with acetone. Sit for a few hours. Decant into flat plate away from any source of ignition as it is highly flammable. Wait about a 1/2 day. Done. Oh, and if you put anything fibrous in the plate, it becomes “magic” too. Simple rolling paper, or regular tobacco work fine. As do things like oregano.

    2) Each plant has a unique character. One was very lively with interesting visuals, another caused more sloth and sleepy happy feelings. Only getting a “mix” hides the unique character of each.

    3) The “5 minute miracle cure”. Leaves can be spread on a cookie sheet and placed in the oven. Turn it on to about 250 F to at most 300 F. In five minutes, retrieve and cool. Done. Worked just dandy.

    4) IF you feed your Iguana fresh leaves, they like it ok. Then they get the munchies and they love it. Then they fall off their perch. ;-) The next day they will wince at the first leaf. Then nibble. Then like it. Then they get the munchies… plop!

    5) With that much available, memory issues manifested quickly. That was when I wrapped up the experiments and “moved on”…

    So guess I need to make that Medicinal MJ posting now… since I’m back “in the mood” ;-)

  40. p.g.sharrow says:

    @EMSmith; at least that might yield a more useful product. I just can’t see making a radioactive pile just for bragging rights and maybe a little hot water. Now a LENR experiment that reduces radioactivity as a byproduct of heat generation might be worth the effort ;-) Solvent extraction of botanicals is fun as well as fairly safe with a little caution. Acetone is much safer then butane and results in a cleaner product then ethanol for solvent extraction. Dry sifting works nicely if followed with several rinses with water. This yields the cleanest medicinal MJ without dirt, vegetation or sap staining. Just way too many things to do just for the fun of

Comments are closed.