We all know how fiction likes to play fast and loose with nanomachines. But really, what would they be capable of in real life, assuming you had an utterly arbitrary amount of them? What would the timescale be? Would you ever be able to shapeshift or use nanomachine-based superpowers, or is that all strictly fantasy no matter what limitations you put on it?

Basically, what separates the IRL potential of nanotechnology in the near future from "NANOMACHINES, SON"?

  • 2
    $\begingroup$ You might look into Neal Stephenson's The Diamond Age en.wikipedia.org/wiki/The_Diamond_Age It covers a lot of interesting things that fiction can do with nanotech. The first things I'd worry about in "real" nano, is the same things I'd worry about with any wearable tech: how do I get it to do what I want, and how do I prevent anyone else from doing whatever they want with mine. $\endgroup$
    – Seeds
    Aug 11, 2016 at 21:42

3 Answers 3



Some arbitrary rearrangement of atoms that's for your designed result rather than part of a metabolism will require energy. There need to be other reactions to provide fuel, and handle waste products. So how do you get a pure result, and deal with available source material?

rare atoms

Natural life uses what it finds and ignores what's useless. But you need catalysts to work on whatever stuff you want to break down, not ignore it! So you need various catalysts, and these often use small quantities of unique and rare atoms. Which happen to not be in the stuff you’re processing!


So, you need feedstock suitable for doing your work as well as the stuff in-situ that you are supposed to be working on.

Your work will produce wastes and byproducts, as well as the result you wanted.


So, the character says “deploy the goo! Develop a program to replace this volcanic outcropping with a marble palace.” And the minions sigh, and haul away the basalt to a dumping site, consuming energy produced elsewhere, taking large quantities of working material sourced elsewhere; then locate calcium carbonate elsewhere. Oh, no prior life to produce limestone deposits? Well, dig through the crust to locate sources of calcium, taking boatloads of energy to do so, taking 100× the structure to make the roots to achieve this. Meanwhile, decide which source of carbon and oxygen to use—if a CO2 atmosphere, grow processing facilities using other raw materials and feed it lots of energy to crack the gas.

So finally you have the feedstocks needed to produce marble. Take that, along with other materials needed to grow the machines needed to do that, to the site. As well as marble deposited in-place, there will be wastes to carry off. And finally the machines need to withdraw and be disposed of.


First, decide what you mean by Nanomachines. The usual rules for Nanomachines involve:

  • Artifices. These are machines, not life. They do not evolve in unpredictable ways and are inherently digital. Seeding a planet with appropriate microbes does not meet your story goals, nor using machines to modifying local biota to make the beautiful rafting trees (as in Titan).

  • Programmed. Dropping the machines has a definite goal. For example, Engines of Creation builds a city and hatchery for a foreign race.

  • Replicating. You only seed a few of the machines. Finding local resources, they create more machines. This whole colony is the factory. This model of self-replication is attributed to Von Neumann.

Now make a few decisions:

  • Goo. Goo is made with the simple rules, with Gray Goo being 'reproduce at all costs'. Aside from the Gray Goo scenario, there are useful models. Asteroid mining programs could be a simple as "create two copies of yourself; fill up cargo with a gram of gold; go to the mining beacon; die", leaving some other tool to gather up the gold mountain covering the mining beacon.

    The alternate is the deeply programmed model. Dense storage in a DNA style molecule would allow complex behaviors with the obligatory gigabytes of cruft for compatibility with old versions of the operating system.

  • Emergent Behavior. Each machine does its thing and the end goal arises from the combined self organized system. For example, surveying nanobots would each try to get a little ways away from any other nanobot before expending itself to make a reading. The mineral map would be the aggregate of each of the datapoints reported by a robot.

    The alternate is networked behavior where the nanobots communicate over wireless links, direct contact, or hub and spoke systems.

  • Single Shot. A single shot of nanobots tries to accomplish a single mission and then die. For example, source gas in Transmetropolitan is good a for a single interview and then goes inactive. The job might take years, for example, sequestering carbon, but only one simple goal is achieved.

    The alternate is a bootstrapping approach, where the nanobots follow a growth and alteration program. For example, they might survey, choose a site, build a solar plant, use the power to make simplier nanobots, create a base, fill it with supplies, and then wait until humans arrive.

Once you have these decisions now you have some questions:

  • Energy Source: Unless you use Deus Ex Machima and supply the nanobots with some shared antimater, you need energy. Most energy is liberated on macroscales, even FOOF provides only a bit of energy in the scales of nanomachines. One pretty much needs to assume either an efficient high-density battery or great energy transmission. Still, most nanobot missions will start with creating a power plant.

  • Armor: The world is a hostile place, full of cosmic rays, radioactive alpha and beta particles, reactive oxygen, and stuff. Nanobots need to repair or defend against the environment in some way. Vacuumn work spheres might work. Fast assembly might work.

  • Complexity: The complexity of a successful intervention with nanobots is staggering. Start with the idea of understanding a landscape solely from single point measurements, go on to how to extract minerals and working materials, and continue onto running a major industrial complex under computer control. Whole industries creating small tweaks and software for nanobots is not unreasonable.

Even if your story is hand-waving over the physics, stating that the handy container of nanobots is the result of decades of innovation and work would be worthwhile.

  • $\begingroup$ Energy isn't really a problem because they could just draw it from all the chemical reactions from breaking various bonds. The complexity thing is interesting, but there are easy solutions to most of that stuff which has been answered with Game Development of AIs ^.^ $\endgroup$
    – Durakken
    Aug 13, 2016 at 0:08
  • $\begingroup$ I disagree. Chemical stored energy comes from organized molecules, usually the result of life. For example, a tree is composed primarily of water and solid air. Energy is used to break apart the H20 and C02 to create much higher energy wood, along with a few trace elements. The wood can be burned to return some of that sunlight energy. No life or past life implies no high energy chemical compounds for the taking. $\endgroup$ Aug 14, 2016 at 2:04

Nano tech is limited, like most other things, by the amount of energy it has access to and the material it can work with. Nanites can manipulate matter at the atomic scale so as long as they have the atoms to make something, they can do the thing you want them to do. This ranges from repairing DNA and cells to creating planets. If they have the material, energy, and instructions to do so they can do it.

As far as Time scale goes. That depends on distance to material and goal combined with travel speed and amount of nanites that there are. That's they could probably travel a few feet per second, lets say 1m/s. Assume all the material you need is at the edge of a 1m^3, have 1 nanite, and have a structure that will take up half the volume of the cube. Find how much material that is in atoms and then multiply by 2 because a nanite can carry one atom at a time and a round trip would take 2 seconds. This gives you how long that structure would take to be made with 1 nanite. For every extra nanite divide by that number of nanites.

Basically, think what you can do in minecraft with just picking up and moving blocks around. That's what you can do, on our level, with nanites.

There is however Femto- level tech which is the next level and likely will be called nanites just because people aren't too good with distinguishing things that are very similar to them, but on our scale their abilities could be very magical looking. Where nanites can manipulate atoms, femtites would be able to manipulate quarks.

Nanite manipulation could cause all types of chemical energy to be released, but once we get to Femtite manipulation we're dealing with nuclear energies and building atoms. Nanites you have to have the atoms to move around, Femtites you need the quarks from the atoms to make whatever atom you want. You couldn't change the mass, but you could change what it appears to us as.

Let's say you want to create an iron statue. Nanites would require that there be iron around to make the statue. If there is no iron around they can't make it and if there is iron in their search area but far away they have to travel to get it. Femtites on the other hand could just pluck quarks from the matter around them, create the particles/atoms needed on the spot, and provide you the statue. Want the statue to be gold? The Femtites just grab the particles needed and add them to the Iron atoms to make them gold atoms... Of course that's assuming that doing this doesn't blow things up, because doing this could cause nuclear explosions since a brute force way of doing this is how nuclear bombs work.

So again, when you think Nanites, think Minecraft more or less. When you think Femtites, think of what would appear to you as magic and alchemy, even though it is still essentially minecraft.

  • $\begingroup$ There's also Picotechnology which is somewhere between nano and femto. $\endgroup$
    – Len
    Feb 14, 2018 at 21:52

You must log in to answer this question.

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