In Stephen Baxter’s science fiction Novel Ultima humans discover mysterious “kernels” which allow high speed space propulsion units to be built. The problem is that these kernels are only to be found 200 miles below the surface of Mercury.

How (in outline) could these kernels be plausibly retrieved from such a great depth in such an environment?

I assume that this is not currently possible with today’s technology, but what challenges would be faced and how might they be overcome using near future technology? Indeed is this even remotely possible?

Assume the active engagement of governments and a virtually unlimited budget. The kernels are 0.5 meter spheres of density similar to the material that surrounds them. They are very tough but might be destroyed by nuclear weapons. Assume that their existence and approximate location are already known.

The idea is for answers to explore the technical possibilities rather than the financial limitations. However financial limitations are a valid concern, so for those who have asked for clarification on what constitutes a “virtually unlimited budget” let’s say $100 billion a year for twenty years, with options for spreading the cost over twice as long if necessary. No quibbling – the politicians have somehow agreed to it (gasp) because the possibilities of the new drive will be so revolutionary.

  • $\begingroup$ The problem with "unlimited budgets" of course is that everything becomes viable. One could just throw money and resources at the problem even right now. But in reality that will never be viable. The most important aspect of this is that resources are scarce. Btw please check out worldbuilding.stackexchange.com/questions/84875/… , this might be more or less a duplicate $\endgroup$
    – Raditz_35
    Commented Oct 20, 2017 at 11:02
  • 6
    $\begingroup$ Unlimited budget is a tricky premise... with unlimited budget just pull Mercury out of its orbit to colder region and dig the hole with ease... $\endgroup$
    – L.Dutch
    Commented Oct 20, 2017 at 11:03
  • $\begingroup$ Not even all of Earth has an unlimited budget. By the time you say "virtually unlimited", you should specify what that means. \$1 trillion? \$100 trillion? \$10,000 trillion? $\endgroup$
    – user
    Commented Oct 20, 2017 at 11:06
  • 1
    $\begingroup$ What's the tech level? How are "kernels" distributed? What time frame do you need? commercial? $\endgroup$
    – user6760
    Commented Oct 20, 2017 at 11:47
  • $\begingroup$ @Raditz_35 This question is very much of interest thx and is related, but was asking about drilling right through a planet like the earth (which was concluded as not being possible) rather than “merely” drilling 200 miles into it in my case. Also I am specifically asking about Mercury which has a different structure to that of earth, is very much closer to the sun, has a much reduced gravity and a long day length making for interesting temperature and radiation fluctuations on the surface. I will adress the budget issue. $\endgroup$
    – Slarty
    Commented Oct 20, 2017 at 13:45

4 Answers 4


From a crater at the poles. The temperature at the poles never rises above 180K. No atmosphere or hydrosphere means no heat transfer. Poles are really the only safe location as Mercury is not actually tidally locked as was previously believed.

Craters at the poles never get sunlight and are hypothesized to contain water ice, in fact this was apparently confirmed in 2012. You can at least assume so for purposes of the story.

Set up some solar panels on mast above the rim of the crater and you should have power. Being this close to sun without an atmosphere, you should not have energy issues.

Then you just dig down. This is helped by the low gravity of Mercury. The pressure underground will be much lower than on Earth and removing material up requires less energy. So you should be able to dig through the crust.

The issue is digging thru the mantle. The 200 miles figure suggest to me you want to dig down to or near to the interface between the mantle and the core. While little is known about the mantle, the core is assumed to be molten, so temperature will be an issue. I think we will have to assume pockets of magma exist, if not entire layers of molten material.

The reduced gravity should in addition to reducing pressures also reduce the rate of convection, so it probably is possible to drill down if sufficient cooling is available. The drill would become surrounded by an insulating layer of lower temperature solid silicates I think. Something like cooling based on liquid metal or metal salts.

After drilling thru the crust, it would probably be impractical to drill further. I think instead the "drill" should just be slightly higher density than the silicates in the mantle so that it sinks down naturally. Then after it reaches its target and traps it by freezing the material around the target, you drop the mass of the drill by removing mass from the top and reducing the mass of liquid metal within. Ie the liquid metal would be both coolant and ballast.

You should then be able to draw the target up. Or really the "drill" would come up by itself and bring the target with it. The head of the "drill" would have to be designed so that it catches the target.

You'd also need some way to detect the target and aim at it. But if you somehow found interesting objects that deep under the surface of Mercury you should already have this information. And probably a permanent base at the pole that just happened to be above one of the objects so it could discover it.

Big problem is really money. The "drill" would obviously be 200 miles long and for sufficient cooling, ballast and mechanical resistance it would have to be fairly thick as well. The head of the drill would also have to be wide enough to trap the object even if everything isn't absolutely perfect, which over a distance of 200 miles probably means the diameter of it should be measured in miles as well. So the "drill" would be absolutely massive.

On a positive side, it would be mostly "just metal" liquid or solid and Mercury should have plenty of metal, especially if you are already building drills capable of penetrating to the mantle. But basically you would still be talking about colonizing and industrializing Mercury for this project.

  • $\begingroup$ I was about to mention the poles as well. One side benefit of a large mining operation on one of the poles of Mercury is that excavated rock can be used to build up a berm out beyond the edge of the crater, creating a man-made "shadow area" much bigger than is naturally available and increasing the area that can be colonized. Basically you'd end up with a man-made solid ridge of "mountains" that can also serve as a base for a solar array (which also increases shadow coverage). In this way you could end up with a lot of usable area. $\endgroup$
    – JBiggs
    Commented Oct 20, 2017 at 14:35
  • $\begingroup$ @JBiggs True. I am not sure if you need bigger area, but this would allow creating a convenient shadow exactly over one of the objects, which might be convenient. $\endgroup$ Commented Oct 20, 2017 at 14:39
  • $\begingroup$ These kernels are small; 500 cm. I am sure the head of your drill is wider than that. Once you get to the kernel, move the bit out of the way laterally. You now have a long tube full of liquid metal which will flow down onto the kernel. Given that the kernel is the density of the nonmetal substrate of Mercury, it is less dense than the metal inside your drill. It will float to the top. Then pick it up, put it on a stick and wave it around in triumph while your team improvises a victory dance. Be sure to cap the drill before all the metal flows out; you can use it for the next kernel. $\endgroup$
    – Willk
    Commented Oct 21, 2017 at 2:02

Like all big engineering projects. You do this in steps.

One of the steps should be retrieving objects from perhaps 200 miles below the earth's surface. Given our current limit of about 7.5 miles, we are a very long way from answering this question.

At that depth, temperatures were about 180 C and even this mild temperature (compared to that on Mercury) was enough of a problem that this was part of the decision to discontinue drilling, as reaching the goal depth would mean 300 C and the drill bit would not work. Mercury is over 400 C at the surface during the day.

Certainly more exotic materials are available for your drill (when money is not a factor). But, having the hole self-sealing due to the rock layers tending to flow back together under the pressure and heat would be a very-big problem. You will reach a point where torque would break any drill we could conceivable make.

You can't just even dig a big crater-style hole either, the rock flow will also make this self-sealing long before your hole is 200 miles deep.

You have to strip mine essentially the whole planet, removing the surface layers into space (to prevent rock flow) -- maybe making a new moon for Mercury.

I recommend you consider Von Neumann machines to perform the strip mining and solar-powered rail guns to ship the material into orbit.


Mercury surface is very hot (430C). Additionally, the Sun radiation as you approach Mercury with your mining mission could cook your ship's electronics and personnel. So, you need to find a way to approach Mercury. The best way is to approach Mercury at such an angle, as to be in its shadow most of the time. Unfortunately, Mercury is small, so I think you could only benefit from its shadow for the last few million of km (I didn't calculate how many) of your approach.

This brings me to the first issue. You need to make sure your ship electronics is resistant to heat. That leaves out a lot of nifty discoveries in microelectronics that make possible today's computing. Your ship electronics have to be simple and extremely robust. Mariner 10 mission approached Mercury, so that much is possible. But, you cannot bring people to Mercury, unless you develop a cooling system and radiation shielding to protect your ship. If you don't, you have to develop materials and new electronics that enable high density information processing and storage at up to 500C, at least.

Then, you have to develop the AI for your ship and robots that would enable you to complete the mission and handle unpredictable situations. I'm not sure if that is doable now, but machine learning has made lots of progress and it is conceivable.

Once you get to Mercury, better land on the dark side (Mercury is tidally locked). You need to build your basis, and bring enough material to be able to start the mining operation. Your space ships can only bring enough material if they have nuclear reactors. If you don't have the technology, better hope you have all that you need to build on the planet. In that case, you need to do a lot of exploration and that might prolong forever your mission.

To dig at 200 miles, I think is only possible with atomic blasts. Assume your hole is similar to those for diamond mining, you should dig an area of at least 100m radius, to have a chance to get to the bounty. That's about 1 billion cubic meters of material, assuming your hole has vertical walls. But, if you dislodge the dirt with atomic blasts, it will most likely be an inverted cone. In that case, assuming a 15 degrees for the cone's angle, which I think is conservative, but possible, the material you need to extract is in the range of ten thousand billion cubic meters. The other thing is that Mercury is quite dense ($5.4g/cm^3$) so you need to break the rock somehow.

You can also do mining on the hot side of the planet. You can use solar concentrators to melt and then solidify the walls of your hole to stabilize them after you extracted the rock.

Looking at the numbers, I think the biggest challenge is getting the rock out. For that, you cannot bring enough equipment from Earth, because Earth has a limited supply of metals. I think you should find enough metal on Mercury, but you should develop an planet-wide chemical industry to separate it and make more machines. That is likely to take lifetimes to finish. One way to do it, would be to make your machines self-replicating.

  • $\begingroup$ Actually Mercury isn’t tidally locked as originally thought, but as the axis has very little tilt and there are craters at the poles which are in perpetual shadow. Although there is a lot of solar energy to deal with, I would imagine that some form of mirror shielding would be within our abilities to construct as well as a small heavily shielded refuge for the crew. $\endgroup$
    – Slarty
    Commented Oct 20, 2017 at 14:34

If we have a virtually unlimited budget, we could just use unmanned craft to push asteroids out of the asteroid belt to such a point that they'll be slingshotted by the gravity of larger planets into Mercury. Leave it to NASA to optimize the exact number, orbits, sizes, and speeds needed to extract the kernels as safely as possible.

  • 3
    $\begingroup$ "Leave it to NASA" does not help someone to write a novel about how NASA did it, don't you think? $\endgroup$
    – Mołot
    Commented Oct 20, 2017 at 11:01
  • 1
    $\begingroup$ You'd need some pretty serious impact force to get down to 300 km depth with an asteroid strike. That's over 10% of Mercury's equatorial radius of 2440 km. You might want to compare the answers to What impact is required for a visible (from Earth) ejecta plume on Earth's Moon, and would the Moon survive? (Full disclosure: My own question.) $\endgroup$
    – user
    Commented Oct 20, 2017 at 11:03
  • 2
    $\begingroup$ @revereche "I find the method more interesting than the exact calculations, myself." The purpose of Worldbuilding SE is to help authors figure such things out. Some people want to go for hard-ish science and real math; others are happy with just general ideas. However, answers that actually discuss a proposal in detail tend to fare better that are little more than just, well, general ideas. $\endgroup$
    – user
    Commented Oct 20, 2017 at 11:05
  • 3
    $\begingroup$ @revereche It will not work... as the hole gets deeper and deeper, the ejecta will fall back down into it. You will just be blowing rocks to pebbles, and pebbles to dust... and the dust to dust, while adding more material. $\endgroup$
    – MichaelK
    Commented Oct 20, 2017 at 12:05
  • 1
    $\begingroup$ @MichaelK I see! The solution would need to be much more elaborate to account for that. Thank you for explaining! $\endgroup$
    – revereche
    Commented Oct 20, 2017 at 12:13

You must log in to answer this question.

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