Building a Reactor | Hacker News Reader

Building a Reactor

usnc.com

131 pointsbeefman30 comments

I worked on this and host it on my site as well where it's not parceled up into bits and pieces. Have to thank Bartosz Ciechanowski. Learned a lot from his code and approach.

https://lvenneri.com/nuclear_reactor_explainer

Somewhat related, a random nuclear reactor generator :)

https://whatisnuclear.com/random.html

Rats. I thought the "Start Company" button would draw up custom incorporation papers for a company building my random nuclear reactor.

> To split atoms and release nuclear binding energy, we need to launch neutrons at the fuel atoms. ... 1 kg of natural Uranium generates just 10 such neutrons per second.

Not saying that's wrong, but my seat-of-the-pants guess would have been ten orders of magnitude higher than this. (Citation?)

https://en.wikipedia.org/wiki/Isotopes_of_uranium

Natural uranium is a mix of 0.7% U-235 and 99.3% U-238. U-235 has a half-life of 700 million years and a spontaneous fission (SF) probability of 7e-11. U-238 has a half-life of 4.5 billion years and a SF probability of 5.4e-7. So on average it takes 6 times longer for a U-238 to decay than U-235, but when it does it has a probability to do it by spontaneous fission that is 7000 times higher. Add to that that U-238 is 141 times more plentiful in natural uranium, and you get that you can completely ignore U-235 when it comes to spontaneous fission.

1 kg of Uranium is about 4.2 moles. Each has 6e23 atoms (Avogadro's number), that's a total of about 25e23 atoms. Every year about 1 in 4.5 billion of those will decay one way or another. That's 5.5e14 decay events. The probability of spontaneous fission is 5.4e-7, so basically 10 million of these decay events are SF. There are 32 million seconds in a year, so I am getting about 0.3 SF events per second. On average a fission event generates about 2.5 neutrons, so that's close to 1 neutron per second.

They are getting 10 per second. They are nuclear engineers and I'm just a guy who knows how to multiply and how to read wikipedia. Most likely their number is correct.

> that's close to 1 neutron per second.

> They are getting 10 per second. They are nuclear engineers and I'm just a guy who knows how to multiply and how to read wikipedia. Most likely their number is correct.

Nicely done! It gets complicated, well beyond my current understanding. However, U238's first decay product is Thorium234, which has a half life of 24 days. Thorium234 in turn decays into Proactinium, which has a half life of 1.17 minutes. There are 16 more decays until finally arriving at Pb206 (lead), which is stable.

These decays must give room for lots more neutrons to emerge.

Those decays don't emit neutrons. Only spontaneous fissions make neutrons. The reactions you're talking about emit alpha particles, beta particles, and gamma rays.

If there are light element impurities in the uranium, or if the uranium is in a compound with light elements, there will be neutrons produced by (alpha,n) reactions on some of the light isotopes.

Some of the neutrons from spontaneous fission will hit other uranium atoms and cause non-spontaneous fission. Maybe enough to amplify it ten times?

Interesting. How much Uranium do we have on Earth?

For all practical purposes the quantity is infinite. The so-called "proven" reserves currently stand at 6 million tons [1]. But the quantity of uranium in seawater is staggering, of the order of 5 billion tons. It is estimated that, if needed, one could extract uranium from seawater at about 10 times the current cost of getting it from mines. That would add less than 1 cent to the price of 1 kWh of electricity generated by nuclear power plants. For comparison the average retail price of 1 kWh in the US is about 17 cents. Of course, nobody is seriously thinking of getting uranium from seawater because there are much cheaper ways to get it from mines. But any talk of uranium lasting only a few decades, or centuries is non-sense.

[1] https://world-nuclear.org/information-library/nuclear-fuel-c...

Depends on what you mean. If you mean "how much is commercially viable" versus "how many kilograms of the stuff are there" you get very, very different answers.

For just "how much stuff is there", we can look at the concentration in the earths crust, and we get about 10^17 kilograms of the stuff (10^15 if you want U-235).

If you just say "Sure, but we can't get most of that, how much can reasonably get extracted at current prices and technology" then you go down to about 10^9 kilograms.

The materials are also difficult to obtain without getting a rooftop guided Champagne delivery. Purchasing prepackaged materials is hard enough, making one is harder still.

Most of Earth's uranium is already engaged in commercially useful activity such as sustaining the geomagnetic field and the geological carbon cycle.

Great looking widgets and graphs.

What tool is used for these ?

Webgl + js. same approach as https://ciechanow.ski/

> The reactor itself, the nuclear core, is not a complicated mechanical contraption. The core has no moving parts or complex mechanisms. It's just an arrangement of special materials, that permits nuclear reactions to occur. There are moving parts to transfer the heat and control the reactions, but it's basically a pile of bricks.

I could say roughly the same thing about a smartphone.

Yes, but a smartphone has billions of microswitches and articulated features. The smallest feature of a fission core is about the same size as a smartphone. The level of sophistication of the tooling needed is miles apart. If we were to start from scratch we could make a fission core long before we could make a smartphone.

And we did, half a century earlier!

However, the smartphone is evolutionary not revolutionary.

Not to pile on but there have been natural reactors in nature, but zero natural smartphones

> pile

I see what you did there

I mean our brains are basically meatsack computers.

Nuclear reactors are definitely simpler. The simplest reactor is just a pure enough and large enough blob of a fissile material. A Turing machine is more complex.

Arguably, the complexity of getting the blob of fissile material should count as complexity. I'd argue that the simplest reactor therefore is the kind you can make with natural material, so a beryllium, heavy water, or graphite moderated natural uranium reactor like CP-1. Even then there was much complexity in getting pure enough graphite.

The article even mentions there were naturally occurring nuclear reactors on earth a while ago.

So it's possible to for a blob of fissile material to just form naturally.

Well, that was 2 billion years ago. Since U-235 has a shorter half-life than U-238, the natural uranium enrichment was far higher back then. And indeed that was a water-moderated reactor. [1]

Some people have postulated that the moon itself was formed in a nuclear fission excursion [2], and/or that there is or was a nuclear reactor in the center of Earth [3]. These are both not commonly believed.

[1] https://en.wikipedia.org/wiki/Natural_nuclear_fission_reacto...

[2] https://phys.org/news/2010-01-moon-nuclear-explosion.html

[3] https://arxiv.org/abs/1308.5934

The Chicago pile was demonstrated in 1942. Transistors weren't even invented yet.

You are incredibly off on complexity

Not sure how any this description could apply to even the simplest microprocessor-containing device. Seems like fabs are sufficiently advanced tech so as to be indistinguishable from magic ... so much so that we forget how amazingly, absurdly, not-simple they are?

Yep you definitely need Unix Theory to build and run a nuclear reactor. I'm saying you're right!

Reply on news.ycombinator.com