Thursday 22 August 2019

How to Live on Other Planets: Uranus

The weirdest of the planets. A blue giant, resting on its side.
Could we find a home in the Uranian system?

Uranus is the fourth-largest planet and the first ‘ice giant’. It orbits between 18.3 and 20.1 AU from the Sun, making it four times more distant than Jupiter and the furthest before Neptune.

The most prominent features are the blue color derived from atmospheric methane and its axial tilt of 97.7 degrees. It also has a ‘surface’ gravity of 0.88g, the lowest of the gas giants, because it is just 14.5 times heavier than Earth.

Uranus is less well known than Jupiter or Saturn. We have less information to work with. What we can say is that it is a huge sphere of frozen methane, ammonia and water ices topped by hydrogen, helium and methane gases. It has a total diameter of 50,700km. 

Pressed between a solid core and thousands of kilometres of ices, we expect to find weird matter like ‘super-ionic water’ and ‘liquid diamond’.

How could we live on this ice giant?

The pressures and temperatures deep inside Uranus are not survivable. 

Unless technologies for surviving those conditions are developed by the time we choose to settle the ice giant, we would be restricted to the upper layers of the atmosphere. It is the smallest and coldest atmosphere of all the planets in the Outer Solar System, so this zone is relatively close to the surface. 

The ‘surface’ in this case follows the convention of being the altitude where ambient pressure is 1 bar. Temperatures in the 1 bar zone are about 76 Kelvin.
The bottom of the atmosphere starts from a mild 320K but a pressure of 200 bars. Volatiles like methane, hydrogen sulphides and water condense into clouds.
Going up, temperatures and pressures fall. The atmospheric composition changes with altitude. 

27 km below the 'surface', pressure is 2 bars and we still see methane clouds at <200K temperatures.

Overall, there is more helium than in other gas (15% by volume compared to 10% in Jupiter, 3.25% in Saturn) while the deuterium/hydrogen ratio is at 0.000055, which is better than Jupiter’s 0.000026 and Saturn’s 0.000017. The methane concentration is a respectable 2.3% as well. As you rise, gases other than helium and hydrogen become less present.

The lowest temperatures in the Solar System exist in Uranus’s atmosphere. It is measured at 50 Kelvin and it is currently unknown why the planetary core’s heat cannot escape to warm up the atmosphere like in other gas giants.

Cold does not mean calm.
The winds in the Uranian atmosphere reach -100m/s at the equator and are strongest midway to the poles at +240m/s. These speeds are measured relative to the rotation of the planet (2590m/s at the equator). We don’t have any detail on the exact wind gradients between latitudes and altitudes.
Uranus also has a ring system. Dust and ices orbit in bands extending from 38,000 to 98,000km above the center of the planet. Compared to other ring systems, there is less dust and more icy bodies in the 20cm to 20m size range.

Within the rings are 27 moons.  

Of the named moons are 13 irregular inner moons, too small to have the gravity to shape themselves in spheres and separate a crust from core. The innermost is Cordelia, 40km wide and orbiting at 49.8 thousand km. The largest of these is Puck, 163km wide and estimated to mass 2,900 trillion tons. It cannot be confirmed, but they are expected to resemble comets and be composed mostly of water ice and some dust.
Five large, spherical moons follow. Miranda, Ariel, Umbriel, Titania and Oberon, named after Shakespearian characters like all the others, range from 471 to 1576km in diameter. 
They have large solid cores and a thick layer of ice on top. All feature huge canyons, crevasses, cliffs and scarps. Tidal locking means they always face Uranus with the same side, and experience 42 years of continuous darkness followed by 42 years of uninterrupted sunlight. The icy surface is darkened by hydrocarbons.
The outermost moons are a diverse group of a few dozen kilometres in diameter, sat in more or less inclined and eccentric orbits that are all retrograde (only small Margaret is not retrograde). Many are believed to be captured asteroids, such as Syncorax. As with the other small moons, they are low density mixes of ice and dust.

Most of these moons sit within the ice giant’s weird magnetosphere. It is weak, weaker than Earths’ by 25%, and the field lines don’t even pass through the center of the planet. 

Once every Uranian day, which is just over 17 hours, the magnetic field connects with the solar wind and allows a stream of protons to funnel down to strike the atmosphere. This could be the reason for the aurorae only recently observed.
Uranuian magnetosphere's evolution
In terms of radiation, Uranus is similar to Saturn. It is safer than Earth and far less dangerous than being near Jupiter.


Living within Uranus is a tough job.

On the upside, the gravity is not too different from Earth’s. The cold atmosphere makes floating with hot hydrogen somewhat easier than on Jupiter or Saturn; 1m^3 of pure hydrogen at 570K would lift 0.38kg if we take the density of the surrounding gas mix to be 0.42kg/m^3. Flying would be no more difficult than trying to maintain an altitude of 9,000m within our own atmosphere, so aircraft would look very similar to our own.

There is a definite need for insulation because of the low temperatures. Heat loss rates would be three times higher than on Earth.

Winds are high, but there is not much evidence of super-storms as on Jupiter and thus any floating or flying habitat should be able to be carried along by them. Some sources state that Uranus's weather system is 'featureless'.

The real concerns are resources, energy and viability.

Uranus is resource-poor. Access to hydrogen, helium and methane is readily available but not much can be done with them. Trying to obtain water would require diving to incredible pressures and is therefore impractical. Raw materials that are required to build or expand a habitat, such as metals and minerals, are simply unavailable.
Art by SolidAhmad.
Worse, Uranus is energy starved. Sunlight is a negligible 3.7 W/m^2. That’s 367 times less than what Earth receives! Any solar energy solution that might be practical at 1 AU from the Sun (perhaps 1kW/kg overall power density) would be worthless out at Uranus’s orbit of 20 AU (2.7W/kg). It would struggle to compete with RTGs at that level of performance.

The winds could be used as was proposed for Saturn, but we do not have enough detail on the windspeed gradients to confirm whether this is practical. If the winds do not vary much between latitudes or altitudes, then the wind-wheels or wind-kites would have to be massive and might not even be able to float.

Nuclear energy is a reliable option as it does not depend on its environment. A permanent habitat would have trouble fuelling its reactors though. There is no accessible source of fissile fuel inside Uranus itself and the densities of the moons does not suggest that they have concentrations of any heavy elements near their surface or at a practical depth. All fissile fuel would have to be transported from the Inner Solar System.
Fusion energy is the game-changer… but not in Uranus’s favour. Deuterium is the most abundant fusion fuel and we can extract it directly from our oceans or any icy body. 
If Deuterium-Deuterium fusion is not yet accessible, then Tritium can be manufactured from lithium and any neutron source. Having engineered solutions at home is a simpler and more straightforward source of fuel than making multi-year round trips to the gas giants to extract Helium-3.

Helium-3 is admittedly the ‘cleanest’ of fusion reactions and it is attractive for use in propulsion and mobile applications, as significantly less neutron shielding is required around a Deuterium-Helium-3 reactor. However, even Helium-3 can be obtained from simply waiting for Tritium to decay. The half-life of Tritium is about 12 years, so large quantities can be produced from a Lithium to Tritium to Helium-3 production chain.

With that in mind, Uranus is the best planet to extract Helium-3 from. It requires much less deltaV to enter and exit the Uranium system than Jupiter or Saturn, and it is closer than Neptune. 

However, the most apt answer to the ‘go to Uranus for Helium-3’ argument is an analogy using vanilla. Vanilla used to be a prized, rare extract from vanilla pods that had to shipped to Europe from the Americas. 

Today, we synthesize vanillin in industrial quantities. It is very unlikely that we would create a space industry around something we can produce at home and have alternatives for it, especially considering that not very much of it would be required to power human civilization. 6,000 tons of Deuterium can provide all of our energy needs every year, and it is only reduced to 1,700 tons with Helium-3.   

That brings us to the question of viability. Why Uranus? It is a fascinating, unconventional planet with a lot to explore, but scientific endeavours do not sustain a permanent population.

If we take settling planets to be an endeavour driven at least partially by comparative economic advantage, then the atmosphere of Uranus holds no advantage and therefore is of no interest. It is a dry, empty and dark desert not worth living in.

Its unique characteristics cannot be enjoyed without supremely advanced pressure- and temperature-resistant technology

Just outside the planet is another matter. Low Uranian Orbit could be a position of interest.

A space habitat orbiting at an altitude of 26,000km from the center of the ice giant (about 450km from the top of the atmosphere) would have a velocity of 14.9km/s. This means that atmospheric gas scooping is feasible with the propulsion systems we have available already.
As described in Low Earth Orbit Atmospheric Scoops, the portion of gases collected by a scoop that have to be diverted for use as propellant depends on the ratio of exhaust velocity to orbital velocity. If the exhaust velocity is lower than the orbital velocity, more propellant must be consumed than is being captured. If they are equal, all gases entering the scoop must be expended as propellant. We want the highest practical exhaust velocity to consume the least gas as propellant and conserve as much of it as possible.

If we want to conserve at least 50% of the gases collected, we want double the orbital velocity as our exhaust velocity.

For Jupiter, this would mean upwards of 140km/s. For Saturn, it is still a huge 50km/s. Uranus is more reasonable with 30km/s required.

While all of these exhaust velocities are achievable, they come at a cost. The thrust delivered by an engine with 140km/s exhaust velocity requires 4.67x more energy than from an engine with 30km/s exhaust velocity. This means that for the same amount of fuel to provide power, you get 4.67x more gas out of Uranus than Jupiter. It is an important consideration when the fuel is likely to be uranium shipped from the Inner Solar System at great cost.

What use are gases scooped from the Uranian atmosphere?
They would be condensed into a liquid, mostly hydrogen, and could be used to refuel spacecraft that aerocapture into the Uranian system, or refill propellant depots supplying space-only (do not land) transports between the Uranian moons. As we will see later, they will be needed for keeping critical infrastructure in orbit around the moons.

The moons

Unlike the other gas giants, the moons of Uranus are disappointing.
The small moons, inner and outer, are assumed to be low density groupings of ices and dust. Frozen water, ammonia and CO2 are abundant in the Outer Solar System, so mining the small moons is not a good reason to go to Uranus.

The large moons have rocky cores that are certain to contain useful metals and minerals. However, they are locked away under hundreds of kilometres of ice. 

They do not have the vast oceans of the Saturnian moons, or the energetic environment of the Jovian moons.

Of some interest however is the moon Miranda.
Miranda is the smallest major moon with a diameter of 480km. It is not very dense (1200kg/m^3), suggesting it is mostly water ice. Surface gravity is a tiny 0.079m/s^2, about 124 times less than on Earth.

What sets it apart from the other moons is that it is unlikely to be strongly differentiated into a core of rock and mantle of ice. Evidence from the Voyager 2 mission suggests that it suffered multiple giant impacts in the past. This would have brought up core material to the surface, to be buried again by ice.

The pressure at the center of Miranda should be about 20 to 25 MPa, which corresponds to 200 to 250 times sea-level atmospheric pressure on Earth. We have submarines that have reached the depths of the Marianas Trench, where the pressure is 1,000 times sea level pressure.  

It is therefore possible, to dig through the entire length of the moon in search of rocks buried in the ice. They can be located from orbit by mapping their effect on the local gravity – as the rocks would be two to three times denser than the surrounding ice, they can be easily distinguished.

The effort of digging through kilometres of ice is not a trivial task, and it is unlikely to be favourable from an energetic and cost standpoint when compared to directly accessing similar rocks in the form of asteroids or rocky bodies.
Ariel is the second smallest major moon around Uranus. It is 1,156km wide, big enough to be completely differentiated into an ice mantle 208km thick and a voluminous rocky core. It is densest of Uranus’s moons at 1592kg/m^3, so it has the greatest proportion of rocks and possibly heavy metals.

The low gravity means that reaching the interface between ice and rock requires handling pressures of only 56 MPa. It is speculated that a subsurface ocean might still exist around the core. It would certainly make moving around and digging out core material much easier… perhaps to the point where it would be cheaper to extract minerals and metals from Ariel than from asteroids.

There is an important comparison to be made between mining resources from the depths of these moons instead of extracting them from asteroids. If it is cheaper to get what you need from asteroids, then no-one will attempt to gather them from the moons, and if this turns out to be the case, then there is no longer a strong argument to settle the moons at all!

Because we can’t reasonably make comparisons based on price decades or centuries into the future, we must find something else. We will look at the energy cost of extracting resources.

Asteroid resources are easy to extract. They do have to be moved however. Altering the trajectory of a portion of, or all of, an asteroid has an energy cost proportional to the deltaV imparted to it.
Asteroid group near Uranus.
Asteroid groups like the Jovian Trojans, Uranus Trojans or the outer solar system Centaurs, are the closest to Uranus and require the least energy to obtain. They are nearly entirely composed of frozen volatiles like ammonia and water ice. Some, such as 10199 Chariklo or 2011 QF99, are huge at 223 and 60km diameter respectively.
It takes about 5km/s of deltaV to move anything near the orbit of Jupiter (5.2AU) to near the orbit of Uranus (19.2AU on average). It then takes another 5.5km/s to spiral down from the end of Uranus’s sphere of influence to a major moon like Ariel, assuming a low thrust propulsion system is in use. Therefore, gathering ‘nearby’ asteroid resources requires between 10.5 and 5.5km/s of deltaV, which is an energy cost of 55.1 MJ/kg to 15.1

The issue with this analysis is that the composition of the asteroids that are ‘nearby’ is very similar to that of the Uranian moons. There is no point in collecting ice at several MJ/kg when you can just scoop it up from a moon like Juliet or Mab.

That leaves the other type of asteroid. The drier bodies, rich in carbon, silicates, minerals and metals. These are mostly found in the asteroid belt orbiting between Mars and Jupiter, at 1.5 to 3.5 AU from the Sun.
The deltaV required to move this material up to our ice giant’s orbit is now 12.8 km/s (for 1.5 AU) to 7.2 km/s (for 3.5 AU). Add in the cost of spiralling in, and deltaV requirements become 18.3km/s to 12.7km/s. This translates into an energy cost of 167.4 MJ/kg to 80.6 MJ/kg.
Keck asteroid capture mission.
Every kilogram of material gathered from asteroids must expend that amount of energy. It is a minimum amount, so larger and larger scale transportation of resources only helps approach this cost. One intangible cost here is the years of travel required to move the resources. DeltaV calculations done so far are based on Hohmann transfers, which usually require the least energy to perform, but take the longest. A transfer from Mars’s orbit to Uranus’s orbit takes 16 years…

How does this compare to digging into the Uranian moons?

A brute force assumption is to calculate the energy cost of melting through a column of ice from the surface to the interior of a large moon. This creates a column of water that can be kept liquid through continuous heating. We have done similar things in Antarctica.

In other words, this is a big up-front energy investment (melting the column) and a continuous cost (keeping the water liquid). Let’s try to calculate these costs.
We will assume a large 10m wide column is dug initially, followed by a huge 200m wide expansion. The reasoning behind this is that settlements are likely to be permanent, or at least last for decades. A large column costs more to create but it reduces the surface area to volume ratio, which means less continuous heating is required. 10m is about the size of a mine shaft for a deep underground mine.

Ariel’s and Miranda’s surface temperatures are 60 Kelvin. If Ariel has a liquid or near-liquid layer around the core, then this temperature rises to about 270 Kelvin. Miranda however is too small to retain heat and therefore its interior temperature is probably around 180K (the minimum temperature for the cryovolcanic activity observed on the moon is 176K according to this).

Water ice has a heat capacity of 2.2 kJ/kg/K. It therefore takes 240kJ to warm 1kg of ice throughout Ariel to its melting temperature, and 264 kJ in Miranda.
Another 333 kJ/kg is needed to transition from solid to liquid. We will also add a 10K margin to prevent random freezing of the water, so that adds 42 kJ/kg.
The total comes out to 615 kJ/kg for Ariel and 639 kJ/kg for Miranda.

The density of ice is between 910 kg/m^3 and 940 kg/m^3 at these temperatures. We’ll use an average of 930 kg/m^3.

The column on Ariel would have to be 208km deep. On Miranda, it is hard to estimate. It is a broken moon where parts of its core could be close to the surface. It has canyons 20km deep, which serve as a good starting point. We can take a conservative estimate and say that two thirds of its radius is sufficient, so 160km. From these figures, we obtain the following data table:

10m column
1.63*10^7 m^3
1.52*10^10 kg
9.3*10^15 Joules

200m column
6.54*10^9 m^3
6.08*10^12 kg
3.74*10^18 Joules

10m column
1.25*10^7 m^3
1.17*10^10 kg
7.47*10^15 Joules

200m column
5.03*10^9 m^3
4.67*10^12 kg
2.98*10^18 Joules

Let’s move onto the continuous heating.

The temperature difference between the water and the surrounding ice dictates how much heat is lost through thermal conduction. This can be reduced significantly by thermal insulation. The insulation can hang off whatever structural support is put into place to prevent the column from collapsing.
A 10cm thick layer of mineral wool that can be made using the rock waste from other mining operations is a good option. Here is the heating requirement data:

47 W/m^2 lost on average
307.2 MW for 10m
6143 MW for 200m

65 W/m^2 lost on average
326.7 MW for 10m
6535 MW for 200m

This is a continuous energy expenditure.

You might be wondering about the energy cost of raising material from the bottom of the column to the top. It will not be calculated because there is a ‘free’ solution: buoyancy. A bag of material can be floated to the top on bags filled with gas. It is one of the advantages of a water-filled column. The other advantage is that the weight of the water pushes back against the external pressure and reduces the structural support requirements.
Now these are all big numbers! How do they compare to asteroid mining?

We know now that setting up a mine on one of the Uranian moons costs a huge deal of energy, and then many megawatts of energy are needed to prevent it from freezing closed. An asteroid mine is assumed to cost nothing to set up from the perspective of Uranian settlers, but asks for many megajoules for each kilogram delivered and years of waiting for the first delivery or for any updates to the required raw materials to be executed.
The initial energy cost is not a deal-breaker on its own. The 10m wide column on Miranda can be created through the work of a 1 GW nuclear reactor and a steam drill for 12 weeks. A giant 200m wide hole can be pierced through Ariel’s icy crust in one year if 118 GW of heat is used.

A useful calculation to perform is the breakeven point at which a hole through Ariel or Miranda becomes the better option over mining asteroids at 80 MJ/kg. It will depend on the quantity of material being extracted and moved up the column.

Here are the results:

10m column
Breakeven in 1 year: 237,348 tons
Breakeven in 5 years: 144,348 tons per year
Breakeven in 10 years: 132,723 tons per year

200m column
Breakeven in 1 year: 49,171,570 tons
Breakeven in 5 years: 11,771,570 tons per year
Breakeven in 10 years: 7,096,570 tons per year

10m column
Breakeven in 1 year: 222,160 tons
Breakeven in 5 years: 147,460 tons per year
Breakeven in 10 years: 138,122 tons per year

200m column
Breakeven in 1 year: 39,826,097 tons
Breakeven in 5 years: 10,026,097 tons per year

These production rates are not excessive. The break-even point for a 10m hole is equivalent to a small mining operation today (~400 tons per day). The large column is representative of the mining needs of a small country. Remember, these tons represent every resource from calcium to titanium. Going by usual terrestrial crust composition, 7 million tons of raw rock extracted only represents 133 tons of gallium or 19 tons of uranium. If the need for these elements is greater than these quantities, then deep Uranus mining is cheaper than moving asteroids.

Note than we have used the less energetically expensive 80 MJ/kg figure and the same 10cm thick insulation for the small and large diameter columns. In reality, there will be a mix of asteroid sources of up to 160 MJ/kg, and the insulation can be gradually thickened. The 200m width of the Miranda mine would have room for 1m thick insulation, cutting the heating requirements to 6.5W/m^2 and the breakeven in 10 years figure to just under 4 million tons per year.

So what does mean for Uranian settlers?

A colony is likely to be based on Miranda if mineral deposits are found near the surface, or on Ariel otherwise.
It will dig chambers at a sufficient distance under the ice to protect it from radiation. Using hot water as a digging tool is effective. The initial source of energy would have to be a nuclear reactor.

The colony would be reliant on interplanetary shipments or resources, people and fuel to set itself up. This could represent years of money being spent with nothing being obtained in return. Unlike Jupiter or Saturn, there would be no ‘novelty’ factor and reduced scientific benefit.

Primary objective would be self-sustainability. Reducing dependence on external suppliers is necessary because the cost of transport is greatly increased by the travel times and distances. We think Jupiter is far away from Earth, orbiting as it does at 5.2 AU. Uranus sits four times further. One factor that would help this colony come into being is the existence of colonies on Saturn. Resources from Saturn would be massively cheaper than shipments from the inner solar system. 

However, Uranian colonists would still want their own supply as it is necessarily cheaper than any interplanetary shipment in the long term.

The path to self-sustainability involves digging these columns, finding a source of power and securing a trade advantage to start becoming profitable.

How can colonies on Uranus power themselves and become profitable? The answer is in the next section.

Uranus Lunar Inter Orbital Kinetic Energy Exchange

Uranus has a number of small moons that orbit at very different velocities. They can serve as the only large-scale source of energy for settlers on this planet.
The concept of a Kinetic Energy Exchange involves accelerating and catching masses between different orbits. More energy is gained than consumed, as it is extracted from the momentum of the moons used as starting and ending points of the exchange. For a full description, read this previous exposition on the concept: Inter Orbital Kinetic Energy Exchanges.

To generate power, we will have launchers accelerate masses into trajectories that intercept receiver stations orbiting moons where colonies are located. The launchers can be self-powered by a nuclear reactors, or by laser beams from the colonies.

We will keep calculations simple by only looking at the two more promising colonies (Ariel and Miranda) as destinations for the masses, and only at Hohmann trajectories.

The first table displays Hohmann trajectories.
Launcher body is the moon where the launcher is positioned. It is assumed to be a coilgun with good efficiency, powered by a nuclear reactor. It shoots projectiles using modest accelerations, so that they can be fitted by a guidance and propulsion system that helps them intercept a receiver accurately.

Orbital velocity is the velocity of the moon where the launcher is positioned, noted in meters per second. The moons further out orbit at slower speeds than the small moons close to Uranus.

Receiver body is the moon where the receiver is positioned. We are interested in Ariel and Miranda. The receiver works like a launcher, but in reverse. Magnetic fields are used to slow down the projectiles, and in doing so, their kinetic energy is converted into electricity.

Receiver orbital velocity is the orbital velocity of the moons Ariel or Miranda, also in meters per second.

Hohmann accelerator velocity is the velocity to which the projectiles have to be accelerated so that they perform a Hohmann transfer that intercepts Ariel or Miranda. In meters per second.

Receiver velocity is the velocity at which the projectiles meet the receiver. In meters per second.

kJ gain per kg is the final result. It is the energy difference between the expenditure of the accelerator and the conversion into electricity at the receiver, measured for 1 kilogram. If it is negative, then more energy was spent accelerating the projectile than was gained at the end. If it is positive, then energy is being taken from the momentum of the moons and added to the colony.

All launchers are assumed to be fixed to their bodies. The effect of gravity on the required launcher velocity is ignored for the small moons, but for the large moons, it means that the projectile is slowed down until it escapes the moon. 

This is why the escape velocity of Umbriel, Titania and Oberon is added to the required accelerator velocity.  

From the results, we see that only projectiles launched from Oberon allow for a gain in energy at Ariel, but a colony on Miranda can make energy out of any accelerator on Umbriel, Titania or Oberon. This makes sense, as shooting out and ‘up’ from the inner Uranian moons costs energy but shooting inwards and ‘down’ in the direction of the gas giant’s pull gains energy.  

The second table is for Hohmann trajectories with orbital velocities around the moons included when relevant.

This time, we have the launchers and receivers orbiting their moons. For launchers around the big moons, we have a boost which accounts for the 71% of the escape velocity. Oberon for example has an escape velocity of 717m/s and then a Hohmann transfer velocity to Miranda of 1,252m/s, creating a requirement for 1,443m/s at the accelerator (it is the root of the sum of their squares). If we have a launcher orbiting at 507m/s, then this requirement is reduced to 935m/s.

The same goes for the receivers. Objects in low orbit around Ariel travel at 388m/s, and those around Miranda at 131m/s. This is added to the projectile velocity at intercept.

kJ gains per kg is now much more interesting. As before, the optimal launcher locations are the large moons closest to Ariel and Miranda, with Oberon coming out on top. Because of the need for the moons to be in the correct positions, this will likely take the form of a reactor charging a large flywheel or capacitor in between launches. All the energy is discharged at the optimal Hohmann launch window. Each 1 ton of projectiles exchanged costs 437 MJ at Oberon but delivers 2,000 MJ at Miranda, for example.

The downside to this configuration is that the energy from the receiving station must be delivered to a colony on the ground. The best way to cross the short distance between the low orbit altitude and the ground is to use a microwave beam and a rectenna to convert the electromagnetic radiation back into electricity at very high efficiency.

This third table is more extreme. It makes maximal use of the possibility to beam energy between moons using lasers.

Instead of transferring energy only in one direction, from a moon in a higher orbit to Ariel or Miranda in lower orbits, the launchers can send projectiles down to the innermost moons and beam back the energy to the colonies.

The advantage is that much greater orbital energies become accessible.

Even better, we can ditch Hohmann trajectories for higher cost, higher reward straight-line trajectories where we first cancel out the orbital velocity, let a projectile drop straight down to Uranus and then intersect a moon at a lower orbit. The relative velocity becomes the sum of that lower moon’s velocity and the velocity of the projectile from falling to Uranus. That is why we now have an extra ‘Gravitational Potential Energy gain’ column in kJ/kg.

For the small inner moons Mab to Cordelia, we assume that a launcher on Ariel or Miranda is used to cancel out their orbital velocities and let a projectile drop. For the larger moons, we assume a launcher is dropping projectiles on Ariel or Miranda.

So, for example, a launcher in low orbit around Ariel would require 5,505m/s to get a projectile to start falling towards Uranus, which is reduced to 5,144m/s if we start in low orbit. If it intersects the orbit of Cordelia, it could meet a receiver at a relative velocity of 10,785m/s. This costs 13.2 GJ per ton but delivers 99.1 GJ.

The disadvantages of this method is that the launchers and receivers need to be much larger to handle the required higher velocities. The process of converting from electricity to laser light and back, and having the beam cross tens of thousands of kilometres, will involve certain losses and a lot of infrastructure, such as relay mirrors orbiting between the moons.

On the other hand, cryogenically cooled lasers are very efficient and positioning mirrors is a small investment compared to the other projects required to start up a colony on Ariel or Miranda. Furthermore, the energy being beamed from the receiver back to the colony can also be used to recharge the launcher, which allows for the limitations of an on-board nuclear reactor to be dispensed with and essentially limitless energy to be unlocked.

Also, energy generation is no longer constrained by Hohmann transfer windows that could be separated by weeks or months, and we could imagine expansions where energy gain is maximized. For example, a captured asteroid could sit at 10 million km from Uranus and an accelerator would only need 760m/s to cancel its orbital velocity. The receiver could be positioned in Low Uranus Orbit and zip around at 15km/s and use atmospheric gas scooping to maintain its altitude. Projectiles exchanged between these platforms could gain up to 334MJ/kg. This is three times better than the best value in the third table.

You can have a look at the calculations here.

We can confidently say that InterOrbital Kinetic Energy Exchanges can fulfil the energy needs of settlements around Uranus. A lot more effort, time and money will have to spent on setting up these exchanges compared to the simple solar panels or wind turbines that are available to colonies on other planets, but once they are up and running, they will be a competitive option.

Uranus Interplanetary InterOrbital Kinetic Energy Exchange
The network of launchers and receivers around the moons of Uranus can be expanded to shoot projectiles across interplanetary distances. It becomes possible for a colony on Uranus to ‘export’ energy to the other planets or more importantly, space habitats in the Outer Solar System.
Example of a kinetic energy exchange between Uranus and Earth.
As we will show, planets that are closer to the Sun than Uranus can receive a lot of energy thanks to the multiplication effect of InterOrbital Kinetic Energy Exchanges, but they have plenty of alternative energy sources. Habitats in the Outer Solar System do not have the same options. Solar power is very weak, nuclear power must be imported at great cost, and the asteroid or comets they work with are cold objects with no intrinsic energy potential. Beamed energy struggles to cross the millions of kilometres that separates them from their closest neighbours, but projectiles can travel indefinitely along certain trajectories.

As before, here is a data table on the kinetic energy exchanges that are possible using Hohmann transfers from an accelerator in Low Uranus Orbit:

We have included hypothetical receiver stations at 3, 10, 20 and 30 AU orbits to represent any colonies not orbiting a planet or moon. In all cases, it costs more energy to deliver projectile to them than it does to generate that same energy in-situ. However, as stated before and especially for stations in the 20 and 30 AU orbits, there may be no alternative energy source.

For receiver stations orbiting planets, we find that the additional boost from low orbit velocity is very helpful. For example, a projectile on a Hohmann trajectory from Uranus to Jupiter would intercept the latter at 3,320m/s. A receiver station in Low Jupiter Orbit is orbiting at about 42,021m/s and would therefore come up to meet the projectile at a combined velocity of 45,342m/s, allowing for a whopping 1 GJ/kg to be gained with each kinetic energy exchange. 

What could be understood from this table is that using simple Hohmann transfers, Uranus could deliver considerable amounts of energy to the entire Solar System.

We can do better. Here is a table where the accelerator is positioned not in a Low Uranus Orbit, but near the edge of the Uranian gravity field:

Because the projectiles do not have to fight Uranus’s gravity on the way out to interplanetary space, the accelerator velocity now equals the Hohmann transfer velocity.

The difference is small when shooting projectiles at the inner planets or those with very high low orbit velocity like Jupiter, but it becomes significant for targets like Neptune. It also becomes a net positive gain to send energy to habitats orbiting at 3 or 10 AU, and only a minor loss for habitats at 20 or 30 AU.

Finally, the extreme case. Non-Hohmann trajectories where we cancel out Uranus’s orbital velocity around the Sun, and with the launcher positioned at the edge of the planet’s gravity field:

The list is shorter because projectiles accelerated in this manner can only fall towards the Sun, and so cannot reach higher orbits. Note that the transfer velocity is exactly Uranus’s velocity around the Sun.

Receiver velocity is the combined orbital velocity of the planet and the low orbit velocity of the receiver station around it. Earth, for example, would add up its orbital velocity of 29,779m/s and its Low Earth Orbit velocity of 7,780m/s, for a total of 37,559m/s and an energy gain of 1.5GJ/kg.

The gains are very high for the inner planets (and Jupiter) and very interesting for stations at 3 and 10 AU from the Sun. They do not benefit from any low orbit velocity because they are not orbiting any other body than the Sun, so maximizing the receiver velocity is very important.

The challenge with this method is that you need to start off with a large mass far from Uranus, or find a way to deliver projectiles from a colony closer to Uranus as well as propellant to keep the accelerator in its orbit.

All calculations can be found here.

What can we take from all this?

The viability of a colony on Uranus might depend on it being able to shoot out projectiles all over the rest of the Solar System. It would be a big energy provider that does not depend on any fuel source and can infinitely expanded.

There should not be any concerns about displacing the moons of Uranus because of this. Slowing down Ariel by 5m/s would yield enough energy to keep our current 20 TeraWatt civilization running for 30 years.

Trajectories and methods described so far are rather simplified and there is much more potential to them. It is possible to imagine that the projectiles use solar pressure or electric propulsion to speed themselves up. Or, they can use gravity assists to reduce the accelerator velocity and still arrive faster at their destination.

The projectiles’ momentum can also be used. A receiver station on an asteroid will absorb the recoil and transfer it to the body it is attached to. This can slow down or accelerate an asteroid, essentially becoming an asteroid moving tool with no reactor or engine of its own.

The projectiles can also contain something of value. They can be shaped into robust containers for something like Helium-3 or platinum.

The value of a colony on Uranus increases as it finds more and more creative way to exploit its position far above the Solar System.

The colony
Do zoom in!
Let’s briefly describe what a colony around Uranus would look like.

Our location pick is Ariel.  

It sits as a network of flooded tunnels inside the ice near the moon’s surface. Two major structures stand out: columns of water extending down to the salty ocean surrounding the core.

The ‘down’ column is smaller, as it exists mainly to send personnel and equipment to the core. The ‘up’ column must accommodate a flow of raw materials and inflated transports that propel them up to the surface, and is therefore much larger.

Flooded tunnels allow for the same buoyant devices that transport raw materials and equipment to used from core to the surface without the need to traverse from wet to dry environments and all the complicated docking requirements that implies.
Ice caves!
The actual living spaces are dry, insulated bubbles within the ice connected by the flooded tunnels. The bubbles are positioned above the tunnels and are equalized in pressure, so there is never any risk of the water spilling into the dry environment. Closed hatches seal off the tunnels to prevent excessive humidity, but they don’t have to be airtight or fight against pressure differences.

Work spaces can be dry, wet or partially flooded according to the colony’s needs. For example, offloading raw materials would be done in a partially flooded space, which would act like a canal dock. Repairing a transport craft could be done in a work space that starts off wet so the craft can be floated inside, and then drains to become a dry space so that the craft rests on scaffolding accessible by workers.

Transport craft carrying raw materials would be slow. They would sink or rise along cables acting as guide rails. Arrestor hooks prevent it from crashing into other transports beneath it in case its inflated balloon pops. Transport craft for personnel are likely to have propellers to speed up the journey and emergency life support in case it gets stuck. Travelling up and down a 208km deep column at 20km/h (10.8 knots) takes over 10 hours each way.

Because of the length of travel, pit stops in the form of solid-walled, surface-pressure habitats that transport craft can dock with would be positioned at regular intervals. These are also good places to form passages to adjacent columns as the network expands. In the further future, they could develop into hubs for workers that want to shorten their commute to the core and live near their place work, and eventually into fully-fledged subterranean ice colonies.

The core and its surrounding ocean is the reason what all this effort is for.

We believe that it is a salty ocean mixed with ammonia and perhaps ethanol. That makes it a good source of hydrocarbons, nitrogen and minerals like sodium and potassium.
The core’s top layers are likely to be cold mud. It can be ‘washed’ in-situ to separate out insoluble components like metals, or it can be delivered whole to the surface to be processed.

While we do not know what ‘core mud’ will contain, we can guess that it experienced heated from radioactive decay and is generally similar to that of the rocky planets, since they formed of the same primordial material, but with a much reduced metal center, perhaps not even fully concentrated into one sphere but scattered as enriched deposits around the central volume. Either way, we are likely to find all the ingredients necessary to build a colony and keep a large population alive. There would be no critical shortage of one element or another.
Digging further into the core would reveal depths better enriched in the more interesting elements like nickel, iron, uranium and platinum.

Most of the rest is worthless though. ‘Wastes’ like silicates would make good fillers for the projectiles used in interorbital kinetic energy exchanges.    


The colony will start big.

This is a necessity. Only in a Solar System where Jupiter and Saturn are colonized and there is active interplanetary trade does Uranus seem a worthwhile expansion. And, because of the implied scale of the rest humanity’s activities off Earth, it will have to quickly grow to a size that meets needs at planetary scales.

The initial investment of digging down to a moon’s core through kilometres of ice and installing giant accelerators and receiver stations that can handle large projectiles at up to a dozen kilometres per second, all while collecting the energy they gather into beams that cross thousands of kilometres, will not make it possible for small, private ventures to try their hand at colonizing Uranus.

It will be less family businesses and the Wild West, more like the Three Gorges Dam and the multi-billion corporate ventures of today.

An Uranian colony would also have to arrive after its potential customers are in place. These include a colony on Saturn and free-floating space habitats in the Outer Solar System. In other words, settling Uranus would have to wait until humanity has expanded to its second gas giant and the asteroid mining business has grown and moved past just searching for metals, and is now trading and delivering volatiles from the comets in far orbits.
And, the advantages that Uranus has over Jupiter and Saturn are relatively slim. Despite all the opportunities described here, they can be done in a lesser capacity by those other planets. A non-Hohmann transfer from the edge of Saturn to Earth would cost 47MJ/kg and deliver 1.45GJ/kg. This is less impressive than Uranus’s 23MJ/kg for 1.52GJ/kg, but the potential for a large multiplier on energy gain is still there.

We would have to wait until that slim advantage is worth more than the huge initial investment mentioned previously. That could be centuries in the future. The obvious danger that comes with describing events that takes place centuries in the future is that new technologies, new social needs or radical changes could have occurred that are impossible to predict and could completely invalidate current extrapolation.
For example, a terraforming effort on Mars could call for vast amounts of water and other volatiles. These would originate from comets far away from the Sun, as they are the easiest to displace from their orbits and smash into Mars… which would make an Uranian colony ideally situated to service them with energy, minerals and metals.
Or, fusion propulsion is perfected and travel around the Solar System becomes rapid and low-cost. There would be no need to compare MJ/kg values for obtaining resources, nor would interorbital kinetic energy exchanges remain worthwhile in the face of deuterium and helium-3 reactors… and thus the advantages of Uranus disappear.

Still, with our current assumptions, we can expect the colony to grow as it finds more uses for its projectiles and new customers appear to make use of them. Greater populations will settle the Uranian moons, which creates an internal market.

Eventually, the ability for the projectiles to move asteroids can be exploited to bring rocky or metallic body to Uranus so that they become the primary source of raw materials at a lower incremental cost than digging new tunnels. This can only happen after the kinetic energy exchange network is in place, so we might see webs of columns and chambers within the ice abandoned after a few decades… that is certainly a unique and interesting setting for a science fiction story or game.


  1. Interesting, thanks for your thoughts. I like the idea of a huge corporate base at the very edges of the solar system. It would make a great setting for a horror story!

    1. Things do get moving when there is money to be made!

  2. An estimate I found of total lithium reserves on Earth is only about 16 million tons. That’s really not much if you are talking about launching interstellar worldships with thousands to millions of tons of payload mass. Lithium-7, the most abundant lithium isotope has a pretty low neutron absorption cross section, meaning you have to burn a lot of fission or fusion fuel, and wait a lot of time to convert it to helium-3. At some point, mining helium-3 should become more practical than making it from lithium.

    1. If you can get the D-D reaction to go that will produce He3 & tritium (H3) the latter will decay to He3 with a half life of approximately 12 years. The D-D reaction is slightly easier to get going than the D-He3 reaction, so I expect that if we get fusion to work at all we will use D-D reactions for stationary power plants where a blanket to absorb neutrons isn't a problem & use the He3 from that for situation where we want aneutronic reactions to reduce the need for massive shielding

    2. @jbperry:
      16 million tons of lithium, converted into 16 million tons of tritium, reacting with abundant amounts of deuterium we can find in our oceans, would yield a total energy of 5.44e24 Joules.

      Even if we extra 10% of all that fusion energy, we would have 5.44e22 joules available. That is enough to power a 15TW civilization for all of its needs for 1150 years. I think that gives us enough for a headstart to solve the problems with the more difficult deuterium-deuterium reaction.

      @Jim Baerg:
      Completely right. That is the best way to use fusion fuels.

  3. While using the interplanetary momentum exchange to generate energy is interesting, and provides a rational of sorts for the colonization, I suspect you're being too pessimistic about other sources of energy.

    Solar is always viable, since you can orbit solar collectors of any arbitrary size in zero G. Singular collectors the size of continents are theoretically possible, but as a practical matter platoons of mirrors or collectors will be positioned until you get the required amount of surface area. Mirrors can be very thin films with little mass, and you could make a case for using solar sails to deliver the colony equipment and then re purposing them once they reach Uranus. The colony's mirror area grows with the colony.

    Beamed power from the inner solar system is also viable, simply by extending the "laser web" concept which has been discussed in earlier posts. Since Uranus has large quantities of Helium and ice. there should also be plenty of 3He and D2 available for fusion reactors. Uranus might become a "pit stop" for fusion vessels moving around the outer solar system, and export fusion fuel to the inner solar system where it is less accessible. Combining the Interplanetary InterOrbital Kinetic Energy Exchange with delivering pods filled with 3He would be an interesting "twofer".

    Finally, in sociological terms, the attraction fo Uranus might be more in terms of isolation to carry out social, economic or religious colonization free from prying neighbours or prosecution and interference from the inner solar system's civilization. This isn't necessarily in conflict with the idea that you need a large organization and effort to set up a colony, think of a 22nd century equivalent of the Mormons trekking to Utah or the modern Falun Gong in China, large, well organized and well funded groups which are capable of large scale operations.

    A "step wise" scenario could occur; initial colonization efforts leveraging light sails for transport and conversion to power stations, building mining infrastructure, extracting and refining 3He and then selling it on the interplanetary market via the kinetic energy exchange system.

    As always an interesting post

    1. Thanks Thucydides, always happy to have you comment.

      I agree with your points. For the colony's own power generation needs, solar power is sufficient, even if you need 271m^2 to gather one kilowatt. I was thinking mostly of ways to justify Uranus from an economic point of view. Historically, settlements have grown by providing services or resources that are needed elsewhere. What service or resource can Uranus provide that is needed elsewhere? I thought up of supplying energy, and eventually we can make use of Uranus's cheaper He3 extraction (thanks to a gravity well less deep than Jupiter's or Saturn's), but I don't know of much else.

      While we can always rely on a sociological or ideological motivation to go to Uranus no matter the cost, I believe it is a 'softer' reasoning. ToughSF attempts to go further than that!

    2. While i agree with the idea of understanding the "hard" sciences, human behaviour and history stands this on its end. The Puritans had very compelling sociological reasons to leave England, but this could not happen realistically until the technological issues of shipbuilding and navigation were advanced enough to make colonization of the new world cost effective. We know some Viking families and factions settled Iceland, Greenland and the New World 500 years before the Puritans, but the Viking colonies were very marginal and only the Icelandic colony survived the advent of the Little ice Age. The desire for new lands, the ability to harvest Walrus for ivory and in some cases, the ability to escape the long arm of the law saw Vikings push their shipbuilding and sailing technologies to the limit.

      I strongly suspect that people will be "heading out" as soon as it is technically and economically feasible, with "waves" of colonization driven by a combination of conditions "back home" (for example the exodus of Europeans in the 1840's due to things like the Irish potato blight and failed revolutions in Germany and other parts of Europe) and the ability to actually leave (large ships capable of carrying hundreds of passengers in "steerage" class). An equal argument can be made for "pull", as farmers came to the New World by the draw of "free" farmland, or the various "Gold Rushes" like California and the Yukon. People want to go, and clever operators work to find cost effective ways to provide passage (for a price).

      One other thing which should not be discounted is there are many cases of people heading out to seek one thing, but ending up discovering and exploiting a very different economic niche. Think of the Spanish heading across the Atlantic seeking spices from the Indies, but discovering America instead. Then the English heading to the Americas to mine for Gold and Silver, but coming up empty in that regard, but discovering furs, old growth timber and farmland instead....The people settling the outer Solar System might come across something that we might not consider. My personal favourite was Issac Arthur pointing out Titan could serve as the heart of a massive industrial and computer park, simply by exploiting the cold atmosphere to allow the Carnot cycle to work far more efficiently. Mining Titan for Nitrogen and hydrocarbons become a very secondary consideration at that point.

      Technology allows you to do more, while social and economic factors give you a motivation to do things in the first place.

    3. @ Thucydides

      If Isaac can sell you the digital colony on Titan because of the Landauer limit then we have more options on Ariel. The 6.2 gigawatts per tunnel is free. If the mining operation is collecting 19 tons of Uranium per year then you can build a 50 gigawatt breeder reactor. More if you can utilize deuterium in fission fusion hybrid. On Titan you could disperse energy in the methane lakes and atmosphere. On Ariel you can sell digital data storage. A few gigawatts can power a lot of archival activity. Backup data does not draw any power until you need to retrieve it. 100 to 200 kilometers of ice is a very effective shield.

  4. In old Venus terraforming discussions, one tired old trope was to "blow off" Venus's atmosphere. Compared to delivering hydrogen from Uranus to react with CO2 via the Bosch reaction (making C + H2O) it's absurdly energy intensive to try to remove the atmosphere. Catching hydrogen tanks from Uranus with regenerative braking systems mean hydrogen delivery is a net energy producer. Of course the scale of the task is immense - about 8% the mass of the Venusian atmosphere in hydrogen delivery - but that's terraforming for you.

    1. I wonder if an Uranus-Venus kinetic exchange can deliver enough energy per kg to accelerate carbon up to orbital velocity, since the Bosch reaction itself is exothermic.

  5. Superb post, I have read completely. Love from India.

    robotic vacuum cleaner india

  6. I only read this for a school project in 5th Grade I'm only 10 by the way it was boring but thanks for the tips! :)