Let's say a group of people in the 1850's had access to all modern knowledge of nuclear physics, but nothing else from modern technology. They have at their disposal resources equivalent to those of one of the major governments of the world at the time (such as the British Empire) Would they be able to make and use nuclear weapons?

The key steps here are:

  1. Mine uranium. This one seems pretty straight forward. There was plenty of mining technology for coal and various metals. The main challenge here is identifying uranium deposits.
  2. Enrich uranium. This is big sticking point. Steam-powered centrifuges?
  3. Construct a bomb. It's a little early for dynamite (that wasn't until 1867) to use to compact a sphere of uranium. Can gunpowder replace it? We had fused cannon balls at that point (the "Star Spangled Banner" references "bombs bursting in air" back in the war of 1812), but I don't know if those are enough.
  4. Launch the bomb. No planes or missiles. Will a cannon work? Little Boy was 28 inches in diameter, which is small enough to fit in the Tzar Cannon built in 1586, but I sincerely doubt it would get enough range. Also, how do you launch a nuke out of a cannon without detonating it?

Bonus criteria: how would you protect the people handling the materials and production from getting dangerous doses of radiation?

Bonus criteria: if 1850 is possible, how much earlier in time could this work?

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    Commented Oct 23, 2021 at 3:42
  • $\begingroup$ In the 1800s, they didn't understand how dangerous radiation was. Probably most of the people who handle the bomb, operate the weapon, or help develop it will die of radiation poisoning. $\endgroup$
    – user4574
    Commented Oct 23, 2021 at 4:52
  • $\begingroup$ Related: nuclear weapons without gunpowder. $\endgroup$ Commented Oct 23, 2021 at 13:57
  • $\begingroup$ Iran and the DPRK are having a hard time with step 2 using 2020s technology - or as close to it as they can get their hands on. 1850s would be like trying to run an AI on 1960s computers - the technology gap is just way too big. $\endgroup$
    – J...
    Commented Oct 23, 2021 at 15:04
  • 2
    $\begingroup$ If you want scientific realism, why not follow the science and drop the idea. Else, why not just write the story using your own steps: Mine, Enrich, Construct, Launch? $\endgroup$ Commented Oct 24, 2021 at 20:28

10 Answers 10


Lots of questions.

In order:

  1. Mine Uranium Ore: trivial.
    1b) Purify the ore to a usable product like Yellow Cake (Uranium oxide):
    Mostly trivial. Assuming you have a nice supply of disposable workers, or you remembered how to make radiation detectors and PPE using 1850's tech

  2. Refine Enrich the uranium. This is the hard point. Uranium metal, or oxide, just does not want to separate. You need to turn your Uranium into a gas, and separate that by mass using highspeed centrifuges or gas diffusion techniques and a lot of patience. a lot of patience.
    But how do you gasify Uranium? You react it with Flourine. Uranium Hexafluoride made from U238 is heavier than uranium hexafluoride made with U235. The difference is only 352 vs 349 g/mol. A difference of 0.85 percent. This allows you to sort them by mass, but the process is slow, inefficient, and only provides a tiny enrichment. You need to cycle the product through hundreds of cycles to get weaponsgrade enrichment. (dozens of cycles for centrifuges, but they are harder to build in the first place)
    Which brings us back to the problem of how your scientists measure the enrichment level without access to modern tools?
    P.S. Did we mention uranium hexafluoride? Working with Fluorine is not for the faint-hearted. That stuff eats everything organic, and even most things metal. Even steel gets eaten, it just leaves a (brittle) passive layer that protects. Making seals or valves or pumps that handle the stuff is an adventure. Again it would be handy to have an unlimited supply of disposable labor, to handle the spills, cleanup, dissolved janitors, etc.

Ok, now you have your enriched uranium hexafluoride. It is "easy" enough to extract the metal and shape it into solid shapes. As long as you are very,very careful about how much of the Uranium you concentrate in one spot, because by now you can cause a micro-fizzling-bomb just by stacking two cans of the stuff too close to each other.

  1. Neither gunpowder nor dynamite come close to suiting your needs for an implosion device. You need very pure, tailored, shaped high explosive charges. Made of several different explosive types, and synchronized down to faster-than-microsecond timing. No, forget about making an implosion design bomb, it will be extremely far out of your technology's reach. Besides, you only have U235. MAking an implosion Uranium device is really, realy hard. The "normal" route is to use you U235 to breed some plutonium, refine that, and then implode the plutonium. That's a whole added onionshell of hightech needed, no thanks.
    Fortunately, a simple barrel-type device, like Little Boy, will work quite well, if a bit inefficiently. For this device, a more conventional explosive propellant will suffice for the detonation trigger.

OK, you now have your (5-to-20 tonne weight) nuclear device.

  1. Delivery
    You will not be shooting your bomb at anyone. The device is too fragile, and too heavy.
    You can not drop it from an airplane, the maximum airplane bomb capacity for that era is: zero.
    You might even have difficulty loading it on a wagon, it's so dang heavy!

Only option: Load it on a steam train, or load it on a ship, and travel towards your target destination. Radio detonation is not an option, so you either need to use a timer or some hardy soul to manually trigger the device.

The earliest time conceivable for construction of an implosion-type device is roughly 16 july, 1945. And the US had to shovel a lot of money at the problem to get it that soon. a lot! The more natural era for developing the bomb would have been late 1950's, early 1960's.

The earliest time conceivable for construction of a gun-type device?.... Possibly as early as 1700. (To start the work, it would be a 50-year++ project that would bankrupt several Empires working together)
The trigger tech is likely steam engines, and the metallurgy that enabled them, to manufacture your centrifuges and/or steamdriven pressure pumps for gas diffusion..
The difficulties would be indescribable, and tackling such a project so early would be more likely to cause the end of civilization(by financial exhaustion), than a working bomb.

  • 2
    $\begingroup$ I'm suddenly reminded of a line from Pantheocide after read your final commentary: Early in the war, at least three economists were reputed to have committed suicide after trying to work out how to pay for everything. $\endgroup$ Commented Oct 22, 2021 at 14:07

Absolutely not!

  • Sure, a gun-type weapon is easier than an implosion-type weapon, to the point that they worried about an accidental detonation in case of an aircraft crash, but either one is "high tech" by 1940s standards. You do not want to handle uranium in a smithy.
  • Centrifuges use uranium in various chemical compounds. They're all nasty. Long before the centrifuges come apart, the uranium hexafluoride or chemicals like that eat through 1850s seals. Or any other enrichment technology without modern alloys.
    As you can see, there is a debate in the comments about what is and isn't possible with the specified level of technology. Let me point out that gaseous diffusion cascades are extremely power-hungry.
  • Firing a crude weapon with a giant cannon might twist it to the point where it might or might not malfunction. Besides, the early nuclear weapons were much heavier larger than period cannon shells.
  • 1
    $\begingroup$ Fluorine and fluorine compounds do not need to be stored in tanks using "modern alloys". Steel or nickel is used, because the fluorine quickly reacts with the inner surface of the tank, forming a passive layer. Having some background in materials science and nuclear chemical engineering, I would actually say that materials available circa 1850, are probably just about sufficient to refine and enrich uranium (using diffusion, not centrifuges). What was not available was the chemical knowledge to design a production chain, and to produce reagents. $\endgroup$ Commented Oct 22, 2021 at 12:17
  • 2
    $\begingroup$ You're missing the biggest one: precision. Precision manufacturing was new at the turn of the 20th century, and you really do need to hit very tight tolerances, even just in your infrastructure. $\endgroup$
    – fectin
    Commented Oct 22, 2021 at 14:28
  • 2
    $\begingroup$ @fectin Can I ask what source you are basing that assertion from? The basis of precision engineering was very much a 19th century innovation. Lapping plates were invented in the 1830s, the Whitworth standard for screw threads was established in 1841 (when "thou" or thousandths of an inch became a commonly used unit), and the micrometre was invented in 1848. Nobody would pretend Victorian engineering is up to modern levels, but their precision was good enough to make pipework that can carry fluorine compounds and uranium raffinate streams. They just didn't know what they would need to make. $\endgroup$ Commented Oct 22, 2021 at 15:03
  • 2
    $\begingroup$ @MadScientist sure, and watchmakers had that level of precision for much longer. But you need that precision on an industrial scale to make an enrichment apparatus (see e.g. L. Dutch's answer), as well as throughout the rest of the process. That industrial-scale precision was an early-1900s American innovation. It's why, for example, the American-built Merlin engines developed more horsepower than the British ones, despite being nominally the same design. $\endgroup$
    – fectin
    Commented Oct 22, 2021 at 15:17
  • 2
    $\begingroup$ @MadScientist - it's worth noting that the loss of UF(6) to reaction with the metals in the pipes was so expensive (even for a US on a full war footing) that they created teflon. They invented it just so that less of their hideously expensive compound got lost in the pipes. 1850 is just before the historical development of the very first man-made plastic, so the inefficiencies, even given the chemical knowledge to handle hexafluoride, would leave the process out of reach. $\endgroup$
    – jdunlop
    Commented Oct 22, 2021 at 17:39

There's one option available if the civilization in your story is at 1850s technology levels for a very, very long time.

The fundamental reason that it's difficult to produce a nuclear bomb is that uranium 235 is hard to make, but easy to use, while plutonium 239 is easy to make but hard to use.

The reason that plutonium bombs are so difficult is that plutonium production always contaminates it with a small quantity of plutonium 240, which has a high rate of spontaneuos fission, and so causes gun-type bombs to detonate as the masses approach each other, releasing enough energy to destroy the device but only a small fraction of the energy of a true nuclear bomb detonation.

However, there is actually an easy, low technology way to remove all the plutonium 240 from a sample of plutonium 239: just wait 10,000 years, as the 240 has a shorter half life.

This suggests a vision of an evil empire who has ruled the land for time immemorial with a nuclear fist, such that no rebellion is imaginable. Deep in the catacombs under their castle, the imperial mages harvest the weapons that were laid out to age by their long forgotten ancestors, and set out new weapons for the long sleep, so that their distant, distant descendents can continue the regime.

You know what they say: the best time to plant a nuke is 10,000 years ago, but the second best time is today!

  • 4
    $\begingroup$ Hard to imagine that in 10k years a better way to make nukes wouldn't be discovered, though. We went from no knowledge of radioactivity to the Tsar Bomb in 60-ish years. $\endgroup$ Commented Oct 22, 2021 at 4:19
  • 3
    $\begingroup$ Minor correction: Pu-240 has spontaneous fission, not fusion $\endgroup$ Commented Oct 22, 2021 at 7:25
  • 2
    $\begingroup$ This would work better if you started with thorium instead. The empire would irradiate natural thorium-232 in nuclear reactors, producing uranium-233 mixed with uranium-232 (half-life 68.9 years). Since U-232 is an intense gamma source, it is normally very difficult to use U-233 to make a nuclear weapon. However, if you wait a few hundred years for almost all of the U-232 to decay, you're left with an ideal nuclear fuel. Pure U-233 has a low critical mass and a negligible spontaneous fission rate, so you can use it in gun-type weapons even with low explosives like gunpowder. $\endgroup$
    – Thorondor
    Commented Oct 24, 2021 at 18:18

I don't think it is possible.

Just to give you one reason: refining Uranium requires centrifuges to separate the right isotope:

U-235 weighs slightly less than U-238. By exploiting this weight difference, you can separate the U-235 and the U-238. The first step is to react the uranium with hydrofluoric acid, an extremely powerful acid. After several steps, you create the gas uranium hexafluoride.

Now that the uranium is in a gaseous form, it is easier to work with. You can put the gas into a centrifuge and spin it up. The centrifuge creates a force thousands of times more powerful than the force of gravity. Because the U-238 atoms are slightly heavier than the U-235 atoms, they tend to move out toward the walls of the centrifuge. The U-235 atoms tend to stay more toward the center of the centrifuge.

Although it is only a slight difference in concentrations, when you extract the gas from the center of the centrifuge, it has slightly more U-235 than it did before. You place this slightly concentrated gas in another centrifuge and do the same thing. If you do this thousands of times, you can create a gas that is highly enriched in U-235. At a uranium enrichment plant, thousands of centrifuges are chained together in long cascades.

At the end of a long chain of centrifuges, you have uranium hexafluoride gas containing a high concentration of U-235 atoms.

The creation of the centrifuges is a huge technological challenge. The centrifuges must spin very quickly -- in the range of 100,000 rpm. To spin this fast, the centrifuges must have:

  • very light, yet strong, rotors
  • well-balanced rotors
  • high-speed bearings, usually magnetic to reduce friction

Meeting all three of these requirements has been out of reach for most countries.

Serially manufacturing an object which can spin at 100000 rpm without shredding into pieces and projecting blades all around is not something one can do with 1850 technology.

You need the technology and the materials.

  • 6
    $\begingroup$ The Manhattan Project did not use centrifuges to enrich uranium. It is also true that the Manhattan Project cost about 2 billion real gold US dollars, which would be about 22 billion debased paper dollars of 2021; no country could have afforded that in 1850. $\endgroup$
    – AlexP
    Commented Oct 21, 2021 at 15:41
  • 2
    $\begingroup$ The project could take 30 years, instead of 3 -- might make it a lot cheaper. $\endgroup$ Commented Oct 21, 2021 at 15:47
  • 3
    $\begingroup$ @SerbanTanasa: By cheaper you mean more expensive? Infrastructure costs are the same, personnel costs multiplied by 10? (And anyway lots of stuff was eye popping expensive in the 1850, starting with all the ultrapure materials.) $\endgroup$
    – AlexP
    Commented Oct 21, 2021 at 15:51
  • 1
    $\begingroup$ @JohnO, plutonium can be only produced in significant quantities by irradiating uranium in a nuclear reactor. And the nuclear reactor requires again fissile uranium $\endgroup$
    – L.Dutch
    Commented Oct 21, 2021 at 18:01
  • 3
    $\begingroup$ @L.Dutch, you can make plutonium with a natural-uranium reactor -- the RBMK is one such design. The problem is that such a reactor is much larger than an enriched-uranium reactor, much harder to design, and much harder to safely operate. $\endgroup$
    – Mark
    Commented Oct 22, 2021 at 1:39

Tac nukes are a thing and 28inches is huge, try 200mm @ 90kg.

But I'm afraid you do need high explosive to reach the critical mass in the shortest time possible, otherwise you'll only get a dirty bomb - enough fission energy to vaporize the nuke but not enough to create a mushroom.

Edit: as pointed out by Zeiss Ikon the the comments, low explosives can be used to accelerate the fragment that will bring the entire fissionable mass to criticality. In fact, the very tactical nukes that I linked up in my answer use this method.

The biggest problem is uranium refining, tho'. Centrifuges Technical means for enrichment are possible within manufacturing precision available in 1850, but the UF6 may be a bit problematic to obtain and handle with the material of the time and in the amounts required for a nuke.
Assuming they have solved the problem of materials resistant to fluoride attack, the separation techniques available at that stage, used alone or in cascade:

As for the mechanical precision available of those times - it's a fascinating history how the humble screw that we take today for granted was one of the essential ingredients for the industrial revolution.

There's an YouTube channel, Machine Thinking, which has a respectable number of clips exploring the precision mechanics through its history; its no wonder for me that a good number of them are dedicated to screws.

  • 3
    $\begingroup$ Centrifuges for uranium enrichment were evaluated and considered not feasible in the 1940s; I don't see how they could have been built in the 1850s. The first practicable centrifuges were built in the 1950s. $\endgroup$
    – AlexP
    Commented Oct 21, 2021 at 15:50
  • 2
    $\begingroup$ With the gun-type weapons possible using uranium (Little Boy style) a longer tube can be substituted for high explosive to allow bringing the "bullet" up to high enough speed to impact the target before criticality can deflect it. The thing might wind up too long to transport on a rail car of the day, but it ought to be possible. $\endgroup$
    – Zeiss Ikon
    Commented Oct 21, 2021 at 16:35
  • 1
    $\begingroup$ If you ditch the requirement for a delivery method, a low-grade dirty bomb smuggled into the enemy's base somehow and just blown up manually (by timed fuse or by suicide bomber) might be just within the realm of possibility... $\endgroup$ Commented Oct 22, 2021 at 16:17
  • 1
    $\begingroup$ @AdrianColomitchi You said centrifuges, so I'm talking about centrifuges. Not enrichment in general. $\endgroup$
    – DKNguyen
    Commented Oct 23, 2021 at 2:00
  • 1
    $\begingroup$ @DKNguyen point taken, I'm going to correct my answer to take into consideration technological solutions for enrichment with the precision available at the time. As for citations about the achievable precision: modern screw cutting lathes, late 1700. Babbage differential engine, early 1800 $\endgroup$ Commented Oct 23, 2021 at 2:09

Just a couple of small notes to round out your thinking to help you either drop the idea or to let you properly represent at least the not-the-fissile-material-itself aspects. Or to help you know what to obfuscate or have major hand waving written in for.

First, handling nuclear materials. As noted in one answer, the phrase disposable workers is a BIG BIG thing. Take 35 guesses at what killed Madame Curie, the person most associated with early investigations into radioactive materials. Ah, just need 1 guess? And bear in mind she only used small samples. She didn't process tons a day for a career. Since your workers would, without changes to "history" on a fairly large scale, so I'm guessing not part of your desired writing plan... since they would do that, they would maybe have very short careers, longer in the mines, yes, but shortening at each step along the way. Since the "along the way" steps get more and more skilled, you'd see the early/swift massive breakdown in health, and outright deaths, definitively affecting your manufacturing silo with the quickest and most debilitating occurring with precisely the hardest workers to replace. Not really a "sentence the thieves to machining materials to a one in one hundred thousandth of an inch tolerance" kind of proposition here, so you'd need to figure a way around that...

On the other hand, but related, we go on and on about the massive economic cost and how it could/WOULD beggar nations. This both is and is NOT a direct concern. First, in the US: the bomb was actually only the second costliest weapon program for us during WWII. Second, not first. And we did both, in spades.

Germany had a program that languished more due to leadership being of the "Keystone Cop" variety. (Like that or not "Nazis-were-gods" folks... not saying you admired them, not AT ALL, but there was and is, a perception that these stumblebums were somehow amazingly on their game when they were really sitting on top of a very competent society and squandered, utterly, most of the actual advantages that gave them rather than capitalizing on them... successes came in spite of them, not because of them.) Add to the stumblebum aspect the utter waste of millions of talented (and stupid, lazy, and very not talented) people, not just in the death camps, or as very low-level slaves before declining to uselessness (to the Nazis), but also tens of millions more, the Slavs in, say, Ukraine, who'd've done almost anything to see Papa Joe dead and gone. (To get a scale of the waste, Alfried von Krupp und Etc. was the largest slaveholder in history owning/operating more than 10,000,000 slaves. That's from his Nuremberg trial. And Germany had many more. But you don't build atomic bombs out of an economy that drove itself in that ruinous direction rather than maintaining itself. But to the Nazi leadership and its underlings fighting to rise, or just to stay alive another day, everything the sun touched was their "commons" to rape and ruin because none of them owned it.)

Even with all that against them, what actually made their bomb project not succeed, more than anything, was that since the leadership had no real concept of it becoming a real thing. (One bomb to kill 150,000 of Papa Joe's soldiers and a few thousand tanks with one blast??? How could that actually be a real thing? There weren't even Norse gods or witchcraft involved so...) Without constant demand from the top, the venal, "kill to gain position" government and army chain of command ensured it would always be something to fail in and end up on a meat hook next to sides of beef (Yes, that was a thing for the Nazis.) for days while someone still had an interest in torturing you. Not really a program with prospects. So, no Bomb.

The British could have run a successful program, and did make important/critical contributions to the American project, people-wise and tech-wise. Why duplicate efforts? Seriously. And with the American effort absolutely impossible for any Axis bombing attack, to say nothing of on the ground military attacks, to reach, it was better to do things in the US. (The Japanese balloon carrying bombs that travelled 7,000+ miles were NOT directed at anything more definable than "the west coast of America" so... not directed at Hanford.)

The USSR also could have, but had no real reason to as conventional efforts were working very nicely and the US was supplying an extraordinary amount of material so their factories could concentrate on the fun things, tanks and rifles and bullets. So not much impetus: NOT "not much ability"...

So, back to the US effort. I'll just bring up one thing to give you the scale of it. It is the SCALE OF IT that matters to your writing. The scale will play out upon whatever economy and population you have, and that will give you cost. Not that cost matters: it is the resources in people mostly, and their skill sets that matters here as neither they, nor their skills, would still be in your economy. If it takes every blacksmith in existence in the nation, then who makes the rifles and bullets and so on?

You have surely heard of Hanford, the eventual factory of the bomb? Quiet, unpopulated western area... In the middle of 1944, it had 45,000 construction workers alone. Not fissile material workers, CONSTRUCTION workers. Think of that. 0 population to 45,000 construction workers, perhaps a soldier or two, actual program workers, oh, and ALL the support people backing them up three meals a day and clothing and so on.

And Hanford was just Hanford. It did not include Oak Ridge or any place in Arizona and New Mexico. Just Hanford.

How would that massive scale work out in doability in your world? Farming done with horses and men, not tractors and combines. Every machinist and so on taken for this project alone. No rifles or bullets except those produced by people who were homeless before your wars. Those people and all their support NOT in your armies themselves, or in support of them. I've read that the US had 16,000,000 men in uniform at the end of WWII out of something like 170-190,000,000 total population, yet no more than something in the 500,000 man range in actual combat, or close to actual combat. Huge armies take even larger support efforts, and could your societies to that and STILL support such a massive drain of people and talent and all their support?

All that is admittedly not likely to be thought of by the average reader. So maybe not a practical problem for you, the author. But... maybe it is.

By the way, fuel oil air bombs, "thermobaric" bombs, could have existed in 1850 with the skills and tech they had. They could have been delivered via balloon, PERHAPS, if someone developed plastics, lightweight plastics that is, much earlier than we did. If oil and its related products (kerosene, plastics) had a one or so generation headstart, maybe two... To give you an idea about balloons, don't think those pretty, pretty fairgrounds things. Think Roswell, balloons launched to monitor the atmosphere for nuclear explosions on the other side of the world, balloons that were 1,000 feet tall once at sufficiently low atmospheric pressure and commensurately sized around, large cylindrical balloons. Possible perhaps in your world. Those would have perhaps been able to carry significant kerosene and some machinery to disperse (mist) it and let that mist fall toward earth, then be ignited by the balloon suddenly flaring bindings, dropping lit materials down into the mist falling above an army or city...

We apparently have achieved 15 kiloton explosions with thermobaric weapons. 75% of the Hiroshima bomb's usual stated blast strength. The Russians/Soviets made one that achieved 39.9 kilotons, or twice the Hiroshima bomb's blast strength, though to be sure, not out of 1850's available kerosene... Admittedly, both from a far more technically advanced economy, but... good chance that's just because that's just the economy in place when people got to thinking on the subject, right? So why not that thinking in 1850-ish?

I'd suggest you go that route. I myself think of it in a fantasy setting. A boy who one day rides a dragon in such a world's wars remembers a frontier barfight in which a lantern is shattered and the fuel spread out into the air in the bar tent as someone swings it at people as a weapon itself, then the mist portion reach a fireplace and the blast killing the "bad guys" but only throwing the good guys, including the young boy, a hundred feet, say, (or twenty, whatever feels decent at the time, eh?) ripping through the canvas, fortunately not splattering any of them on a tent pole or boulder (outside), then the slamming punches of flame following out through those rips in the tent before the general blast vaporizes the tent and all concerned still in its immediate area. That boy, now a man, leading the dragon arm of the army he is in, having large more or less waterproof sacks made sized to the dragons who'd carry them, lantern fuel collected from the besieged capital (large) city, then the force (oh, "wing"... why not, right?) taking flight one night and wax-covered holes in the sacks eaten away by the kerosene, the contents sieving out into the air, falling a thousand feet breaking into a mist to a reasonably large extent though maybe most lumping back together into more of a rain, the riders, task done, flying off wondering what it was all for, then some noticing this man directing his dragon into the falling rain of kerosene above the center of the enemy camp and having his dragon breath the best blast of flame he's ever breathed... Death on a scale no such medieval society would have spent much time imagining. And sacrifice at the very center.

(I picture it via a young magic user as well, either way, and I hate dragon stories, so I spend more time on that avenue, mentally.)

Did the little bit above seem reasonable? Not Hemingway brilliantly written there, just... reasonable? In a medieval tech-level society without examples like successful atomic and fusion bombs to inspire them to find other ways of achieving something massive?

If it did, then why not in your world of 1850's tech instead of nuclear weapons? It just needs someone to notice that a mist of kerosene is incredibly more explosive than one would think at first. And that's not too much of a stretch as powders would easily have been a very present thing in such a society. Grain dust in huge storage silos, gunpowder in its making and storage, even powdered metal could exist as a noticeable thing in your world, and any powder from flour to gunpowder and beyond either direction can explode. Coal dust in mines (Benxi, Liaoning: 1,500 dead). Look up grain explosions. Flour would definitely make a mist as it fell, and not gather into rain-like "droplets."

I'm saying there are alternatives that can still give you horrendous and massive explosions to work with. Beats hand waving any day!

But if you need it to be nuclear, not much help.


I'm going to tentatively say that it is possible, but with a few caveats.

The 19th century was a time of rapid technological development. Even if someone magically went backwards in time to give someone the blueprints for a nuclear weapon, it would likely take decades to develop a weapon with the technology of the time (by which point technology would be noticeably post 1850s), and this process would require other technological innovation specifically to facilitate the project (just as the Manhattan project did in its time).

The notion that people could have a sophisticated enough understanding of nuclear physics that they could build a bomb with 1850s technology is very, very problematic. For instance, how was the neutron discovered? Do they have a quantum mechanical model of the atom? Do they have particle accelerators? And if you answered yes to any of those questions, why didn't they learn how to make other more useful things with that technology?

So, with those caveats out of the way, let's work through some scenarios:

Time Travel

Your people have been given detailed instructions that cover the theory and engineering requirements of a nuclear weapon, from the future, some advanced alien race, or divine revelation. Everyone just knows that it all works, and the government is very, very keen on seeing it built, no matter the cost. To me, your initial question seems to be hinting at a scenario such as this.

In this case, an 1850s nuke might be plausible. Development may need to begin a little earlier (say, 1830s). Uranium would be mined, but your chemists would know how to extract it from ore and enrich U-235. You will then need to enrich it, using gaseous diffusion. But your engineers will have been told how to do that too (including required enabling technologies), so they likely could. This would be the most expensive and time-consuming part of the operation.

To build the bomb you would want to use a U-235 gun-type configuration, where you propel two sub-critical uranium blobs into each-other. How difficult that is will depend on how much you enrich your uranium. This is mostly a matter of cost.

Building a delivery system is the tricky part. Your bomb will be big and cumbersome, so you'll probably just put it on an armoured ship, send it towards the enemy fleet and have some brave soul on board detonate it. In principle you could set up some timing device, but given how expensive these things will be I doubt whoever is paying for them will want to risk it. I don't think an aerial delivery system would be feasible. You could also use the bomb in traps or sieges, where it is simply moved into position and detonated at the appropriate time.

Then again, the first powered flight occurred in 1852 using a blimp-like airship, so you might consider allowing that technology develop further in your story and dropping it over the enemy. But your airship will still be vulnerable to enemy fire. Delivery of the weapon then comes down to protecting the airship until it arrives where you want it.

Alternative history, with some geological modifications

If you're happy to modify the natural world in some ways you can make the natural abundance of U-235 much higher. One way to do this is to make the Earth itself younger. U-235 has a half-like of 700 million years, so if you shift the entire evolutionary history of the earth back 700 M years you will have double the current natural abundance of U-235 (~1.4%). Going back 1.4 billion years gives you double that again (~2.8%), and 2.1 billion years gives another doubling (~5.6%). This still doesn't make it easy to build a bomb, but the uranium enrichment process will now be much simpler.

Explaining how your society came to realize building a nuclear bomb is feasible is still problematic, but you might be able to hand-wave that away by arguing that they built dirty bombs first and nuclear research was stimulated by that.

Different physics

If you allow for new fissile isotopes that exist only in your universe, things get much easier. Suppose U-238 is now weapons-grade fissile material. Problem solved. Building the bomb is now mostly about developing chemical extraction processes for uranium.

Since building the bomb is now much cheaper your people may be willing to take more risks with delivery systems. But I still don't think you'd want to risk launching a nuke from a 1850s cannon, so moving the bomb into close proximity with the target may still be the goal. Again, I see naval battles a major application of this technology.

Strictly 1850s, as it was in our history

Sorry, but no. It would require too many technological innovations in too many highly specialized places and too many accidental discoveries for something like that to make any sense at all.

I don't care about feasibility, just make it happen

In this case, we're back to the Time Travel scenario, except your bomb was developed via a string of extremely improbably chance discoveries.

  • 1
    $\begingroup$ Welcome Matt, nice and thorough first answer. Enjoy Worldbuilding. (From review). $\endgroup$ Commented Oct 22, 2021 at 6:33
  • $\begingroup$ Nice answer. Another scenario that might work, depending on OPs goals, is something like the Dark Tower. Due to some catastrophical event in our future, human kind lives like in the 1850s, but are left with remnants of what we have today. $\endgroup$
    – johnny
    Commented Oct 22, 2021 at 10:39

TL;DR Use elements that occur in strongly radioactive elements by default to avoid centrifuge requirements, build a breeder reactor using a neutron source.

If you drop the uranium use, you may have a better chance. Naturally occurring radioactive elements without need of isolating isotopes such as radium and thorium were already available in pure form around 1910 (not 1850, so you are still lacking 60 years of technological development). They were also available in your timeframe in their natural ore, but I'm not sure that is sufficient. Together with aluminium and beryllium, you can use them to construct a neutron source which you can use to enrich all kinds of radioactive materials (note that those metals were ridiculously expensive to refine back then). You can then try to make a bomb using the enriched radioactive materials.

You will likely lose a lot of workers to radiation poisoning.

You should probably check with a nuclear engineer or someone with a similar skillset. I (obviously) have not tested this approach.

Some references:


You don't use implosion devices but the simpler (less efficient) gun design.

Absolutely doable with an artillery, actually it's been done:

nuclear cannon


One possible issue is refining, but steam power, the less efficient gaseous diffusion enrichment method, and accepting a dirtier bomb w/ lower yield will probably get you there.

The only limitations really are in regards to the difficult mathematics required, but maybe you can spin up some babbage engines.

  • 6
    $\begingroup$ In the days of the Manhattan Project, computers were women who computed. $\endgroup$
    – AlexP
    Commented Oct 21, 2021 at 15:38
  • 1
    $\begingroup$ True, but they also used analog computers. $\endgroup$ Commented Oct 21, 2021 at 15:46
  • 4
    $\begingroup$ Gaseous diffusion requires uranium hexafluoride. UF6 is nasty stuff -- it corrodes most metals, it'll destroy rubber or leather valve gaskets in short order, it reacts with water, and it's quite poisonous. Producing it requires gaseous fluorine, which, in the mid-1800s, was still busy killing off any chemist who tried to isolate it. $\endgroup$
    – Mark
    Commented Oct 22, 2021 at 1:34
  • 1
    $\begingroup$ The problem with gas diffusion is: You need to manufacture a suitable semipermeable membrane using 1850's tech. None such was available. Additionally, you need the ability to pump and keep heated vast quantities of Uranium Hexaflouride gas. How do you do this without electric pumps? $\endgroup$
    – PcMan
    Commented Oct 22, 2021 at 14:26

The existing answers mostly overlook a real problem, although hinted at in some.

There are two big issues, not just one.

First, you need enough enriched material that a bomb is possible. The difficulties in producing this have been explored in detail in other answers.

But the second problem is that you need to keep the impending explosion contained for the tiny interval of time required for a few percent of the material to fission. That's a real challenge too.

A fission reaction in an above-critical amount of uranium/plutonium, occurs extremely quickly. The problem is that whether you use implosion (hollow sphere), or gun (rod entering sphere with a hole in it), you have to assemble that mass of material, and hold it together during the tiny amount of time it takes. Otherwise your material will melt, and blow itself apart and the main result will be dirty fuel spread out and a bunch of hard lethal radiation, but not the explosion you hope for.

This is why atomic bombs require such precision engineering. The imploded sphere needs the implosion shockwave to be almost exactly symmetrical, the gun method needs its precision hardened steel casing, but both need their shapes engineered and calculated to hold the material together during the intense conditions of that very initial stage, or it wont work.

And that kind of materials skill just didn't exist in 1850. There wasn't an obvious way available to them, to design a device that assembled subcritical mass into supercritical mass, and held it together afterwards, as needed, because precision engineering of that kind requires an entire industrial skills base that didnt really exist AFAIK.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .