BPM 37093 is an object believed to be the nearly exhausted core of a dwarf star, with about 110% the mass of the sun. It is especially notable because it is believed to be composed of diamond — a solar mass of diamond. The “value” of this object is incalculably high because it breaks the underlying assumptions of scarcity in economics — at 363,000 times the mass of the Earth, if we were to somehow possess this object, also referred to as Lucy, from the Beatles song — diamond would become by far the cheapest substance, cheaper than gold, aluminum, or even sand.
So let’s go get it, right? No matter how expensive it may be to get Lucy, surely the eleventy quintillion dollar value justifies. But Lucy is fifty light years away, impossible to even reach within the scope of human history, let alone return with a payload. Even if the transit problem could be solved, Lucy is the dead core of a star — good luck taking off from the surface of a solar core once you’ve packed up the gear. There are a thousand other problems we don’t even know about.
So whatever figure you choose as Lucy’s value, however astronomical, if you’ll pardon the expression — it’s immaterial. The value of an irretrievable object is zero. It’s important to keep this in mind when we see beautiful infographics like this, from Wired:
This is a graphic representing a number of asteroids whose composition we think we have a handle on. The area of the circle is the value of the asteroid, with the color representing type — the reddish color indicates a metallic asteroid, which obviously will be the most valuable, at least in the near term.
Zooming in on the graphic, we can see this:
This figure represents the “cheapest” of the asteroids represented in the graphic, with a value of $1.12 qt. What is “qt”? “qt” stands for quintillion, i.e., 1,000,000,000,000,000,000. A billion billions.
“Embracing the suck” is an American military term which means “don’t kid yourself about how hard the mission is going to be — face it head on.” These dollar numbers are mind-boggling. But let’s remember the lesson of Lucy, the diamond forever in the sky — an irretrievable object is valueless. But 206 Hersilia isn’t 50 light years from Earth, it’s actually about a fourth of the way between Mars and Jupiter, a place we’ve sent nine probes to, or to beyond that. Notably, the Dawn mission recently visited Ceres and Vesta, two objects in the neighborhood of Hersilia. The Rosetta mission actually landed a probe on a cometary nucleus.
But landing a probe on a target body is only the first step in a thousand needed to make the mission economically viable. Here are a few of the unsolved problems that follow (for the purpose of simplification, I’m going to focus on platinum as a target ore):
- The actual mining
Every proposal I’ve read stipulates robotic mining. But fully robotic mining (as strictly distinguished from teleoperated mining) is not even something we can do on Earth. The Stillwater Mine, one of the most advanced platinum mines in the world, still employees hundreds of human beings to run its operations. Entirely new techniques and devices will need to be improvised and invented for low-g and low material cohesion mining conditions. While much of this can be computer-modeled and predicted, any mining engineer will tell you that there is no substitute for observing conditions in the field.
It seems unlikely that a survey even as detailed as remote-sensed seismography of an asteroid would yield information good enough to risk the investment of a mission that could ultimately fail because the gear couldn’t for example, perform rake mining instead of drilling a shaft. While many aspects of an asteroid mining operation may be quickly automated, in the near-term, asteroid mining missions will need to be manned missions.
2. Ore processing
If you think that the actual mining is difficult in zero-g, that’s just beans compared to ore processing. A common method of refining platinum ore relies on foam flotation, a process whereby the ore (having been pulverized) is suspended in water, with the platinum particles adhering to air bubbles. In a zero-g processing environment, the minerals would need to be separated centrifugally, perhaps with the entire vessel being part of the centrifuge, as I discussed here.
At this point, the ore is typically smelted to remove the nickel and sulfur impurities, which are significant. The spacecraft would need a high temperature to perform this smelting, which in turn would need a great deal of power.
Why not simply return the unprocessed ore? Because that will involve transporting mostly cheaper, and probably uneconomic minerals. Mining silicon or even iron from space is unlikely to pay the bills. Refining the ore to a high degree of purity may not be necessary, but given the high transport costs associated with this venture, the less waste you’re transporting, the more profitable it will be.
3. Returning the payload
The Apollo programs returned a sum total of 382 kg in payload from the Moon. The Space Shuttle returned a 100,000kg to Earth (that includes the mass of the craft), and the SpaceX Starship is slated to deliver 100,000kg of payload — to Mars. It seems like a high mass payload return (the higher the mass, the more value the mission) is a solvable problem, but with the retirement of the Shuttle there is no existing stack that has demonstrated heavy payload return.
Why does payload, per se matter? What’s the difference between returning a craft that has a 100,000kg payload, and simply returning a craft that weighs 100,000kg more? Payload matters because spacecraft must be load-balanced, especially when considering retropropulsive landings. 100,000kg of pure platinum occupies a mere 38 cubic meters, a cube 3.36m or 11 feet on each side. Imagine that cube slipping a foot to one side while you’re trying to land the craft — the manned craft.
There’s no law that says that the payload must be returned in a single landing, but for every landing, there has to be a launch, increasing the costs and lowering the margin.
The Shifting Sands of Commodity Economics
Let’s stipulate that we’ve solved these Herculean problems, and we have trucks offloading 100,000kg of platinum at a sunny, south Texas landing pad. The platinum spot price is about $26,000 per kilo — that works out to a cool $2.6 billion dollar take for the mission! At that level, the mission is bound to be a barn-burning economic success.
But hang on — 100,000kg is roughly 60% of worldwide annual production, about equal to the production of South Africa, the largest producer. Increasing the worldwide supply by 60% is likely to drive the price significantly downward.
The name for this situation in Economics is price elasticity of supply (PES). The price of platinum can be volatile. In May 2008, the spot price hovered around $2000 per troy ounce ($66,666 per kilogram). Seven months later, it was at $882 ($28,356 per kilogram). A massive change in the supply that would come with a big mission like this would likely trigger a strong PES response.
There are some big asterisks here — I’ve focused on platinum, which happens to be at a very low price point in recent history. And while platinum has plunged, palladium has risen. So focusing on a different ore (and perhaps a different asteroid) can solve these economic problems. The point I’m trying to make is this:
The payload necessary to make an asteroid mining mission economically viable may so large that it would push significantly on the PES for the ore.
If this happens, and a mission is only marginally economic, the PES could push it back into the red. Aluminum was once more expensive than gold — the cap at the top of the Washington Monument was out of it. But as improved refinement processes and industrialization took hold, the price declined from $500/kg to $40/kg in five years.
Of course, the other way you push on this equation is by making the missions cheaper and faster — but that is something that can only be done with repeated missions. If initial missions are not economic, the curve will never reach that point.
The Other Side of the Argument
One possibility that blows a great hole in the side of all of this doom-talking, and that’s that the asteroid ore could be vastly higher in purity than we find terrestrially. Platinum is extremely rare in the Earth’s crust, but it is believed to be far more common in the mantle far beneath the surface. It’s possible that these asteroids are composed of mantle material, and the level of processing and refinement is vastly less than is necessary on Earth. It’s possible that chunks of pure platinum may exist as rocks sitting loose on the surface of the asteroid.
In a case like this, it really may be possible to simply send a robot with a scoop arm to grab it up. It might be possible to survey a large enough sample to simply strap a rocket and navigation package to get it back to low Earth orbit, where it could be packed away and returned to Earth.
This kind of mission, cheap, easy, fast, small — is enabled by a very high purity ore asteroid. This leads to a conclusion —
We need a very highly detailed survey of a great number of metallic-type asteroids. To this end, the U.K.-based Asteroid Mining Corporation is proposing two Asteroid Prospecting Missions, APS1 and APS2. APS1 will create a map of the composition of a number of objects, while APS2 will physically visit the best of the results produced by APS1, and gather detailed surface information in order to determine how to extract the minerals.
The other point to consider is the flip side of the commodity price argument. China has substantial holdings in rare earth minerals — were those minerals to suddenly become unavailable for political reasons, or even reasons of warfare — the prize for solving all of these difficult problems might suddenly make sense.
Given everything we’ve covered here, the entire industry of asteroid mining is dependent on good answers to two questions:
- How rich is the ore in these objects?
- What is the availability of the ore?
The best answers are “extremely rich and lying loose on the surface”. The worst answers are “like here at home (or worse) and found primarily in subsurface deposits”. APS1 and 2 could shed some light on these questions.