HomeLatest ThreadsGreatest ThreadsForums & GroupsMy SubscriptionsMy Posts
DU Home » Latest Threads » Forums & Groups » Topics » Environment & Energy » Environment & Energy (Group) » The Structure of Molten U...

Wed Feb 6, 2019, 04:16 AM

The Structure of Molten Uranium Dioxide.

Last edited Sat Feb 16, 2019, 02:51 PM - Edit history (1)

The paper I'll discuss briefly in this post is this one: Molten uranium dioxide structure and dynamics. (Skinner et al Science Vol. 346, Issue 6212, pp. 984-987 (2014))

The famous "Elephants Foot" produced by the melting of the Chernobyl nuclear reactor in 1986 consists of a mixture of steel graphite, silicates produced by melting concrete and about 10% uranium dioxide which also melted during the event.

The thing has been photographed, even though a range of a meter, the exposure to radiation emitted by it will be fatal in about 5 minutes.

Here's the photograph:

I hope that guy in the picture didn't stay there very long.

The photograph is grainy because the film was being exposed to radiation when the picture was taken.

I personally am interested in what happens when nuclear fuels in a liquid state, since liquid nuclear fuels whether in solution, such as being explored in various kinds of "molten salt" reactors, or (more to my interest) liquid metal fuels, particularly low melting eutectics made using the remarkable properties of plutonium offer certain advantages that solid phase fuels cannot match. One value of these nuclear fuels is that they allow for on line in process separation of fission products. This is, by the way, what happened at Chernobyl. Much of the cesium and other volatile fission products boiled away, contaminating a large area. It is seldom noted that this kind of thing can also be contained, thus lowering the available inventory of fission products for leaking into the environment, while recovering valuable radioactive elements including, but not limited to, radioactive cesium. Radioactive cesium, like many other fission products, utilized in a controlled way, can solve many intractable chemical pollution problems in air and water.

Some time back, in this space, I offered a long involved "thought experiment" showing how the use of radioactive cesium to clean up the air might work: A Scientific Rationale For Pursuing New Immobilization Forms For So Called "Nuclear Waste."

Uranium dioxide has an extremely high melting point, thought to be around 2,865C. Precise measurements are difficult to obtain, since the hot oxide will interact with the materials containing it, reducing the purity of the sample while destroying the container, as happened at Chernobyl.

To avoid this problem, the authors of this paper, which was designed to utilized a very innovative approach. They "levitated" the sample in a stream of flowing (unreactive) argon gas. From the paper's supplementary information:

The aerodynamic levitation method involves floating of the sample on a on a 99.999% purity Ar gas stream inside an Ar filled chamber (21). The sample is then heated from above using a 400W CO2 laser (Synrad Firestar i401). Temperature was measured at the top of the sample using a Chino IR-CAS pyrometer (0.7-0.9 μm waveband). The liquid UO2 emissivity value of 0.84±0.03 was used in the temperature correction (2) and the error in temperature is estimated to be ~2-3% due to a combination of temperature gradients in the upper part of the sample and chamber window transparency corrections. A brass sheet was placed in front of the area detector to absorb any uranium fluorescence (L-edge ~20keV), while passing >80% of the elastically scattered x-rays (standard corrections were used for attenuation).

Their goal was to learn about the structure of molten uranium dioxide.

From their introductory text:

Nuclear power from fission currently accounts for about 10% of the global electricity supply. Compared with burning fossil fuels, nuclear power has prevented ~64 × 10^12 kg of CO2-equivalent emissions since 1971, corresponding to a saving of 1.84 million air pollution–related deaths (1). Because the majority of currently operating nuclear reactors use either UO2 or mixed oxide fuel (typically 90% UO2), understanding and predicting the behavior of UO2 at extreme temperatures is of great importance to improved safety and optimization of this low-carbon electricity source. Although no experimental structure measurement of molten UO2 has been previously reported, some physical properties measurements (2, 3) and several literature molecular dynamics (MD) models do exist for molten UO2 (4–9). These models, which are often parameterized from solid-state properties, have large differences in their melt structures. The U-O bond length rUO, for example, varies from 1.9 to 2.2 Ĺ between UO2 melt models (4–9). This structural uncertainty results in molten UO2 models with differing physical properties, such as viscosity and compressibility, that are relevant to reactor safety. The present x-raymeasurements, by contrast, find a precise value rUO of 2.22 T 0.01 Ĺ (at 3270 K), which provides a new tool to test the validity of liquid UO2 models.

Here's a nice picture showing the apparatus (as a schematic) and some x-ray data:

The caption:

Fig. 1X-ray diffraction measurements of UO2.
(A) UO2 x-ray structure factors. Above 1300 K, the high-Q region (Q > 12 Ĺ−1) contained no structure. The setup diagram (inset) shows the incident x-rays passing through the collimator (1) and the hot sample (2). The temperature was measured with a pyrometer (3), and exhaust gas was filtered (4). The view through the exit window shows the sample loader (5) and beam stop (6), which absorbed exit window scattering. (B) X-ray pair distribution functions DX(r), generated from the patterns in (A). Dotted lines are from the Yakub MD model (6); red lines (3270 K) indicate the liquid state. As an approximate guide, these UO2 x-ray diffraction patterns consist of pair contribution weightings of 73% U-U, 25% U-O, and 2% O-O (at Q = 0).

What happens when UO2 melts is that the coordination number of oxygens bonded to uranium decreases.


The caption:

ig. 2UO2 coordination numbers and bond lengths.
(A) U-O pair distribution functions [rTUO(r)] and running coordination numbers [nUO(r)]. Open circles are nUO(r) from the x-ray measurements at 2100 K (black) and 3270 K (red). The above 3-Ĺ U-U correlations also contribute to the measured nUO(r). (B) U-U pair distribution functions [rTUU(r)] and running coordination numbers [nUU(r)]. In both (A) and (B), light gray and light red curves are the rT(r) and the thicker, darker curves are n(r). The dashed or unbroken curves are from the Yakub MD model at three temperatures: 2100 K (dashed gray or black), 3000 K (unbroken gray or black), and 3270 K liquid (unbroken red or pink). The direction of the arrows indicates increasing temperature. The dotted red lines are the n(r) curves from the refined MD model. The temperatures chosen are either side of the lambda transition in the hot solid (2100 K and 3000 K) and the stable liquid state (3270 K). (C) Measured rUO(solid circles) and rUU (open circles), normalized to the 300 K value. The dotted lines are the Yakub MD model. Red circles (3270 K) indicate the liquid state. Liquid UO2 number density at 3270 K is 0.0593 Ĺ−3 (2).

The authors perform some molecular dynamics calculations and structure determination


The caption:

Fig. 3Molten UO2 structure measurements and MD simulations.

Diffraction measurements (red), Yakub and refined MD models (black dashed). (A) The thin, light gray line on the upper pattern is the raw measurement, whereas the smoother red lines correspond to the solid red lines in (B). The inset shows times for 50% of bonds to be broken in the solid or supercooled liquid at 3000 K (light blue). (B) The solid red lines were filtered by Fourier transforming with an r-dependent modification function to reduce unphysical high-frequency noise (13, 19), whereas the solid light gray line shows the unmodified transform. The lower dashed and dotted curves are the UU, UO, and OO partial contributions to the x-ray pattern. (C) Slices from the refined MD simulation (~15 by 12 by 3 Ĺ) showing U-O polyhedra, above and below the lambda transition (2100 K and 3000 K) and in the liquid state (3270 K). The bottom slice shows the UO6 drawn in black and the UO7 in light blue (11).

Here are some brief concluding remarks from the paper:

Portions of the hot solid and liquid UO2 MD simulations illustrate the large oxygen disorder above the lambda transition and the different UO6,7 coordination species that predominate in the melt (Fig. 3C). The structure and optimized interatomic potentials for UO2 allow for accurate atomistic multiscale modeling. The x-ray data are important as an end-member benchmark for models of multicomponent systems, including corium melts and high-level waste glasses (11).

Cool, (or hot) paper.

By the way, I'm opposed to the use of "waste glasses" to dump so called "nuclear waste," since radioactive nuclear materials, albeit limited in the amount that can accumulate by the Bateman Equations.

We ought to be utilizing these materials to solve otherwise intractable problems. Of course, our current generation is too stupid and too paranoid to understand as much, but future generations, one hopes, needing to clean up our mess, dumped on them as an expression of our contempt for the future, will understand as much.

Speaking of the future:

One of the labs to which my son applied for his summer internship was the lab of this paper's authors, Argonne National Lab. He was, however, offered a job at Oak Ridge National Laboratory - his first choice - which he accepted. It's neutron work.

I'm jealous, but very proud of him.

I wish you a nice day tomorrow.

9 replies, 468 views

Reply to this thread

Back to top Alert abuse

Always highlight: 10 newest replies | Replies posted after I mark a forum
Replies to this discussion thread

Response to NNadir (Original post)

Wed Feb 6, 2019, 04:53 AM

1. Something for the morning, when I have coffee.

Well done to your son, though.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to Dead_Parrot (Reply #1)

Wed Feb 6, 2019, 09:27 AM

2. It's nice to hear from you again.

I've wondered where you were.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to NNadir (Reply #2)

Wed Feb 6, 2019, 09:56 AM

3. I had... Stuff to do.

I'm glad you are still here. mate. Hope you are OK. Will argue with you after coffee.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to NNadir (Original post)

Wed Feb 6, 2019, 12:26 PM

4. So I had to wiki UO2, out of curiousity ...

DU being favoured for the uranium component due to its low radioactivity.[7]

Who knew ?

Seriously, though I'm amazed at the number of interesting electrical properties of UO2 being investigated.


Looks like thoria is still a better choice for crucibles, which is the only application I'm likely to come into contact with.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to eppur_se_muova (Reply #4)

Wed Feb 6, 2019, 10:15 PM

5. Uranium nitride is the really cool binary compound of uranium.

These days whenever I think of a solid nuclear fuel, it's usually a nitride.

The melting point of UN is roughly 3110 K. (Melting point determination of uranium nitride and uranium plutonium nitride: A laser heating study (Nunez et al Journal of Nuclear Materials 449 (2014) 1–8)

Cool paper. I have the PDF, but not the graphics, which I pull out of the HTML files to which I don't have home access.

The interesting thing about nitride fuels involves the fact that some fission product nitrides don't form, for example Cs3N under these conditions, whereas other nitrides decompose at differing temperatures - giving a different separation profile - and also that they decompose (often) into the elements. This paper talks about Pu/U mixed nitrides, and the melting is done under an N2 pressure, but the system behaves quite differently in the absence of N2 pressure.

PuN decomposes, apparently at a lower temperature than UN in the absence of nitrogen pressure. That, I think, is a very useful property.

Many of the actinide oxides, carbides and nitrides are extreme refractories.

I've scanned a wonderful monograph from 1986 on "Advanced LMFBR Fuels" which has a wonderful discussion of oxides and carbides, and nitrides.

One feature of the nitrides is their very high thermal conductivity. This is an obvious advantage over oxides and carbides.

I've been interested in learning more about their corrosion resistance; from what I can see not much is known.

My son's research project last summer in France concerned polymer derived ceramics and as I live vicariously through his work, I spent some time in libraries looking into these compunds. I considered the potential of these polymer derived ceramics as "breathable fuels," composed of uranium (or other actinide) nitrides. (I somehow suspect that something similar is going on in Bill Gates' "Traveling Wave" breed and burn reactor. It's supposed to be proprietary, so I don't know.)

I'm thinking along completely different lines than breathable solid phases, more into composites with improved resistance to fracture.

One "problem" with nitride fuels in some people's minds is the generation of 14C. That doesn't bother me a bit, I must say, since 14C is a wonderful isotope because it has a much lower neutron capture cross section than the two stable isotopes of carbon. Some people want to use 15N nitrides, but that involves isotopic separations. Not worth it, I think.

Beautiful and very interesting chemistry, these nitrides. You know, I never thought much about nitrides when I was a kid; probably they were an afterthought if even that. I played with azides here and there, and even deliberately decomposed a few, but nitrides, no.

I now own a few books on them, mostly built around their ceramic properties, some focusing on the non-metal BN, SiN, and oxynitrides.

Cool materials, very cool materials. I wish I'd learned about them decades ago, but alas, I didn't.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to NNadir (Reply #5)

Thu Feb 7, 2019, 06:44 PM

6. Aluminum oxide nitride (ALON, in commerce) is a promising optical material.

It's what the blastproof windows in HUMVs are made of. I'd love to get ahold of a sample and try grinding a thin, lightweight telescope mirror from it. There's been a large amount of use as wideband windows for drones etc. but I'd settle for optical "seconds" -- wouldn't matter if it was opaque, I just want that high specific modulus !

I think I was originally a bit biased by the ease of hydrolysis of some nitrides -- then I learned what really stable ceramics some are. Of couse if you go back far enough, NO ONE knew about that.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to eppur_se_muova (Reply #6)

Fri Feb 8, 2019, 12:05 AM

7. Oh, I know what you mean. I first realized this when studying nuclear fuels...

...but when my kid started his education in Materials Science - and I set out to follow what he's doing - a whole new world opened up.

Of course, nuclear fuels are some of the most demanding materials in the world, and when I learned that the nitrides of the actinides were remarkable refractories with high thermal conductivity, it blew my mind.

The nitrides of the other elements are also fascinating.

It turns out that one of my son's professors got his Ph.D (and a post doc) in Shuji Nakamura's lab. Nakamura's Nobel was for just one compound, GaN.

(It's a remarkable story, Nakamura's life; he almost didn't get into college because of his fascination with playing volleyball. I recommend his Nobel Lecture for sheer amazement at a strange career.)

I mean GaN is still a subject of huge research.

It made it possible to include blue light in LEDs, which is why we can now have white LEDs.

The nonmetal nitrides are also amazing, boron nitride is an amazing material, as are the silicon nitrides.

For me though, the class of nitrides that really blow my mind completely are Barsoum's MAX phases. (Some MAX phases are carbides as well as nitrides.) They are the most amazing materials, a mixed bag with the properties of ceramics and metals, machinable ceramics actually. I found out about them before my kid went to college, and I spent a lot of time talking about them in a general way with him; and I'd like to believe that had something to do with him choosing materials science, although I also recall that there was a time when they were in Junior High School that I found them, both my boys, expressing fascination with hydrogels.

It turns out that as I was scanning Barsoum's monograph, my son was actually sitting in a room with Barsoum himself, unbeknownst to me, since he was applying to Drexel's materials science department. He was accepted there, but chose another university where he got an offer he couldn't refuse - his first choice in any case.

There are boron and silicon nitrides that are polymer derived ceramics that are essentially polymers with huge ternary structures, with many of the thermal insulation properties of hydrogels. Full circle for my kid, I think.

For me though, man, like you, I never took them seriously - I'm sure I didn't consider them at all (most of my career has been around medicinal chemistry, proteomics and peptides in any case).

Most of us aren't really trained to think about the inorganic chemistry of nitrogen, but it's exceedingly rich, and I'm a little sad that I didn't really think about it until late in life.

Sorry to ramble, but it's really remarkable, and it's nice to run across someone else who appreciates it....

Reply to this post

Back to top Alert abuse Link here Permalink

Response to NNadir (Reply #7)

Fri Feb 8, 2019, 02:06 AM

8. Heck of a story about Nakamura in the book "Brilliant!".

In no small way did this man change the world.

Reply to this post

Back to top Alert abuse Link here Permalink

Response to miyazaki (Reply #8)

Fri Feb 8, 2019, 09:13 PM

9. Thanks, I'll check it out if I have time. It's one of the most romantic stories in science...

...Japanese marginal student obsessed with volleyball doesn't get to get a Ph.D., gets a job as an industrial chemist in a small firm, works on an impossible project, gets pulled from it, gets sent to Florida, gets dissed by graduate students, continues to work on the impossible project on his own time, gets obsessed, makes the thing, finally gets a Ph.D., makes his company rich, gets next to nothing, sues, gets a lot of money, wins the Nobel Prize and moves to California, all for making blue light.

It really doesn't get much better than that.


Reply to this post

Back to top Alert abuse Link here Permalink

Reply to this thread