Monday, October 10, 2011

Lunar Mining [Part 4]: Eclipse by Reality of the Economics of Lunar Mining


This is our fourth and final article on whether or not it makes sense to attempt to exploit the resources that are available on the moon. When Team Phoenicia started working on our rover design, we thought to ask whether or not we ought to outfit the Victory at Pangaea – our rover – with sensors tailored to help a nascent lunar mining operation kick off. To figure this out, we dug into the costs and economics of mining the moon. These posts that we have been writing are touching on what we found and our very basic conclusions.

That said, we have tried to write this post multiple times. We tried cobbling together several technologically plausible scenarios for how lunar resources could be extracted and sent on their merry way to the markets of earth. We attempted to directly walk through the terrestrial process and illuminate what might be done on a lunar mining operation. After each iteration was written, we examined the resulting product. The articles meandered and lost focus. They laid bloated and indigestible on the screen. They simply did not work. So we sat back and looked once more at what are the real issues are for a lunar mining operation.

In the end, there are really three issues that must be addressed if any one entity will ever exploit lunar resources. The first is whether or not it is viable under the law to find and prove a lunar resource. The second is whether or not it is economical to extract from and refine that resource. Finally, it is whether or not it is even remotely viable to move those raw or refined resources back to a market.

These all combine, like some Japanese anime robot, into a single question. Is it possible to mine the moon and turn a profit with the above problems? Let’s examine each of the three above and see what we find and whether or not they actually fit together for a profit making venture or if they are merely smoke and mirrors…or at best premature.

Finding…and Proving

It’s an interesting dichotomy that exists here for Moon mining schema, it may actually be very easy, or relatively so, to find a resource on the moon for exploitation. Remote sensing has become a wondrous beast these days. Satellites and aircraft are picking up mineral resources all the time now. In fact, aircraft found sources of rare earth element ore bodies that are worth exploiting in Afghanistan that may help that poor place find another path than the one what it is on. Similar technologies to this can be used on the moon and have been. Mineral maps are updated every single time that there is a new orbiter placed around the moon. Their resolution is getting better and better. So long as the regolith is not obscuring the underlying ores, or giving a false positive, orbiters may be able to pick out the target ‘mother lodes’ long before a lander or rover touches down. The search need not be as erratic as it can be on Earth.

The flip side of that dichotomy is not discovering the resource, but proving that it is what your lunar prospecting bots say it is. Consider: your rover that you send up must take five cores that are over 15 m (50 ft) long/deep for examination. Each of these is around 100 kg (220 lbs). These must be examined from top to bottom to verify that the mineral ore is present in quantities claimed. Mining companies then use an independent, trustworthy party to do that assessment.

Stop and read that again. An independent, trustworthy party. There are very few independent parties that will be willing to build their own rover with the necessary instrumentation to examine those cores. If there were, it would mean that they might violate the second part: trustworthy. They would have too much of a vested interest in making the lunar mining endeavor work. That means that the cores must be returned to earth for examination. What does that entail?

To get back samples to the earth’s surface from the Moon takes 2700 m/s of delta V (or change in velocity) if you use aerobraking, if you don’t it’s higher. Note: lithobraking is highly frowned on. While this is highly contingent on what propellants would be used, let’s use what Team Phoenicia has planned: RP-1/LOX. This would require, with the assumed losses, that the dry mass of the rocket and payload be 10% of the total rocket launching from the lunar surface returning to earth. Let’s be extremely generous and state that the payload is 5% of the dry mass of the rocket. What is the total GLOW – Gross Lift Off Weight (erm, mass) – needed for 1000 kg of lunar core sample payload to be returned earth? If the 1000 kg is 10% of the dry mass and the dry mass is 10% of the GLOW? Then we get a rocket massing 100,000 kg at lift off from the lunar surface…

So to deliver that sized rocket would require multiple Falcon9 launches. How many? 50+. At $55 million each…Oh. Oh my. Wipe that. What about FalconHeavies? At $80 million each? Uh. So obviously, we just busted the bank just proving the source. It should be noted that even this unoptimized design is still around $2.75 billion for the F9 and $560 million for the FH, just for the launch costs.

Interestingly, as an aside, the amount of material returned by the Apollo program was in the same range…and cost around $100 billion (2011). Now, the Apolloans did a lot more than merely dig up core samples and the costs of our core sampler rover have not been included, but the worst case scenario is still an order of magnitude cheaper than the Apollo program. The 1000 lbs of material cost $100 million/lbs. With the core bot and SpaceX, this will cost $5.5+ million/lbs.

There are ways to tweak this problem. The propellants can be swapped to hydrogen or methane for better mass fractions. The mongo rocket could be broken up into five rockets to allow for multiple (!) rounds of funding. NASA or some other space agency may be willing to buy the core samples, if legal, from the lunar mining operation. It may be possible to manufacture LOX on the lunar surface from the regolith. Etc. Even so, just proving the resource would still be at least the cost of the bringing a mine from nada to fully operational on earth, from conception to discovery to proving to refining to even down to delivery. And there are several more steps before the moon resource is even a viable mine. This is only proving that the mineral ores are actually viable … if they were readily accessible.

However, let’s make the assumption that the magic $500/lbs to LEO rockets happen soon and can be used for lunar mining? That would still leave us proving a lunar source costing more than opening an entirely new terrestrial mine. What about $100/lbs to LEO? Um. No one is talking that price range now or in the foreseeable future.

In some ways, we ought to call a halt here. We hit such a huge iceberg for this titanic mining venture that it ought to sink. We want to make sure that the lesson gets driven home. Let’s use the $100/lbs to LEO model and take a look at the mining process refinement that would need to take place.

Refining the Process…and the mineral!

For rare earth elements or platinum group metals, the refining process must be tailored carefully to the ore in question in order to extract the desired metals. The laboratories that this must be done in are here on earth. There is no way that this can be done on the Moon without placing a manned colony there and even then it would be extremely difficult to do. Sadly that remains even more uneconomical. However, the dominating cost for working out the details is the amount of material that must be moved back to earth to be worked on.

The answer we got back from the mining folks? 20 tons.

A half ton was as much as the cost of the opening a terrestrial mine. 20 tons now puts us, well past the order of magnitude in cost for transport using rockets than what is required to open a terrestrial mine.

Even so, let’s continue.

Just Get Me to the Church, Get Me to the Church…erm…Market on Time.

Ok. So, it ought to be obvious that working with a rocket even with $100/lbs to LEO is not going to work here. You would need the previous step to cost $10/lbs to LEO to have a snowball’s chance in the sun to make a lunar mine remotely viable up through proving the process, barring tweaks.

The Sixth Way

Might there be another way? Besides the pleasure of possibly making Robert Heinlein smile in his grave, we talked with the folks that are building the United States Navy their whopping big EM cannon – railgun! – to ask a few pointed questions if it might not be possible to at least fling back the refined metals (preferably) or ore back to earth. In digging around, we found that there were two systems being actively developed that might have applicability.

The first was the railgun effort headed up by the United States Navy. This is an attempt to build a new 200+ mile range weapon for the navy. Some herald this as the return of the battleship because surface ships would be able to engage with guns at ranges only aircraft and missiles could before. Whether or not this would be the return of massive surface ships is only tangential to our interests though: we need a highly energetic propeller of mass from the lunar surface to LEO in the form of an EM gun. A railgun most definitely counts as that.

The second system we looked into that is being developed is the EMALS. This is a linear induction motor being developed for the Gerald R Ford class of nuclear aircraft carriers. This is intended as a replacement for steam catapults that have graced American carriers for decades now. The EMALS system recently test-launched an F/A-18E fighter at 20 metric tons (plus or minus and see test below). While not as fast as a railgun, it definitely can drive mass.

Given that there were two systems in development, we thought to seek out the manufacturer of each and find out what they thought. We spoke with Thomas Hurn of General Atomics about their work for the US Navy on railguns and the linear induction motor for the EMALS catapult system for launching naval fighters off the new Gerald Ford class aircraft carriers. Thomas was delighted to work out the numbers for us, but even before he started, he stated that the linear induction motor was not what we wanted. We needed a railgun for any attempt to lob back material from the lunar surface using an EM gun.

Railguns have come a long way since the mid 1990s when I was actively following them. Back in the day, they had significant problems. The power switching for high energy applications was not up for the task of large slugs at high velocity. The rails themselves would arc and erode, as the slug would be launched. The equipment was enormous and unwieldy. However, times have changed and technology has marched on. So much so that the US Navy is looking forward to moving from a lab environment to a shipboard system within the next decade.

General Atomics has a railgun that fires regularly fires at 2 megajoules. The system, named Blizter, masses between ten and fifteen tons. They use it regularly enough that they can and do fire it very frequently. The technology involved is scalable. The technology is even modular so that chunks of it can be dropped in placed as needed for more power and, through that, velocity.

The technology of Blizter is scalable enough that the USN and General Atomics tested a newer, larger railgun here recently. This is not a mere 2 megajoules railgun, but rather a 32 megajoule beast. The vetted public video of the test firing is below.

Consider what a 32 megajoule railgun could do. Kinetic energy is defined as one half the mass of the slug times the square of the velocity. If we had a one kilogram slug, that railgun could lob it at around 8,000 m/s. That’s more than enough umph to get a 1 kg slug from the moon’s surface to low earth orbit. In fact, its enough to get a 16 ½ kg slug back to LEO if aerobraking were used.

The state of the art railgun is enough for being used as a mass driver. However, for lunar mining the railgun we need will be significantly larger. What about one for one ton slugs?

Thomas stated that the technology at this point was relatively linear for building up capacity in the energy to mass ratio, although understandably, he didn’t reveal the exact numbers. After all, its for the USN’s new uber cannon after all that is intended have a 200+ mile (300+ km) range. However, given that Blizter is somewhere between 10 to 15 tons, that gives us a base number to work with if we can figure out the energy required for one ton slugs.

Thomas was kind enough to provide a ballpark number. That number was in excess of 2 gigajoules. That’s enough for us to try to extrapolate a minimum system mass. That would come in around 1000 times more massive than Blizter…which makes it as massive as a USN Ticonderoga class cruiser by itself.

While being awesomely scalable, it would literally require hundreds of FalconHeavy launches to get to the Moon. A one ton slug of pure platinum, our upper bound for a first approximation of economic viability, would be worth around $55 million. Just to cover the costs of the FalconHeavy launches, it would take four plus slugs for a single launch with equipment and that would just cover the equipment for transport back to earth and not the refining.

In the end…

There is no way, even with exorbitant PGM prices in excess of $50,000/kg that it is remotely economically viable to mine the moon. Proving the source alone would break the budget for mine development when compared to terrestrial sources. Getting the mining process worked out, even when not taking into consideration the very unique lunar conditions, will also break the bank several times over. The cost of transport dominates the equation and that will not change sufficiently enough to make the endeavor viable.

Continuing to claim that it is viable and worth pursuing at this point is … presenting as much of a chimera to potential backers as suggesting that mining Helium-3 is. It is a hue disservice to us all. All it takes is a potential backer to consult a mining professional to discredit the position. Then, more than likely, the backer will be averse to working with many, if not most Moon projects. Perhaps someday in the future lunar platinum group and rare earth element mining will become viable, unlike He-3, but…not now. Definitely not now or even in the foreseeable future.

In wrap up, Team Phoenicia’s rover, Victory on Pangaea, will not be sporting a instrument package that is designed around helping a lunar mining operation get underway. Alas.

PS. Yes, this was the short version of the article.

PPS. This applies to asteroid mining as well.


Hop David said...

Helium 3, rare earths or platinum aren't the only possible resources on the moon or asteroids.

Water is a very valuable resource. Propellant high on the slopes of earth's gravity well could make space travel much less difficult.

Not only could orbital propellant make the moon and near earth asteroids more accessible, but also the nearly 200 billion dollars of orbital assets we have in cislunar space.

Charles F. Radley said...

Sending materials from the Moon the the Earth with chemical rockets is not viable. But with a lunar elevator, the economics become very attractive. also, there is huge market potential in space, using lunar feedstock to manufacture solar power satellites at the Lagrange locations, then drifting them to geosynchronous Earth orbit.

A lunar elevator can be built today with existing materials at fairly low cost.

Will Baird said...

Potentially, Hop, water would have some $, but whom are you delivering it to? the ISS? Is it a big enough market? Don't underestimate the costs of filtering out the regolith and whatnot.


Actually, no. It is within the technical capabilities of the US or any other space faring nation. Kevlar will work as a tether for a lunar elevator. However, its not within the economic capabilities of the US. Or probably world. For a single kevlar 1 cm dia cable to the lunar surface from the L1, you need over three million kg. You need almost 200 Falcon Heavy flights for that.

And with folowup flights every month or so. That cable is going to be cut on average every 3 months, if the NIAC study is right.

200 Falcon Heavy flights is an Apollo level of commitment, minimum. And the comment from the Augustine commission was that the reason we're not flying SV's is that we can't really afford them.

Additionally, its a bit questionable if SpaceX can deliver 200 FHs in a reasonable time frame. They have only flown the F9 how many times?

Hollister David said...

"but whom are you delivering it to? the ISS?"

Um.... No. Drinking isn't the only use for water.

"Propellant high on the slopes of earth's gravity well could make space travel much less difficult" evidently needs elaboration.

The major barrier to economic space travel is Tsiolkovsky's rocket equation. See Mf is a Mofo

Water can be split into hydrogen and oxygen, one of the higher ISP chemical bipropellants.

Given propellant depots at various locations, inter orbital tugs would have lower delta V budgets, Propellant mass fractions would be from 25% to 60%. Strong, single stage reusable vehicles would be doable. Delivering communication sats and other orbital assets would be much easier.

Lunar resources could enable more than just reusable inter-orbital spacecraft. They night make reusable earth to orbit vehicles possible. If an upper stage didn't have to endure the extreme conditions of an 8 km/s re-entry, re-use would be much less difficult.

To sum up lunar water might make economic, reusable space craft possible.

"Don't underestimate the costs of filtering out the regolith and whatnot. "

That remains the open question. LCROSS told us there's at least a few percent ice at the polar cold traps. The elevated CPR from Chandrayaan-1's radar suggests ice sheets at least two meters thick. It would be worthwhile to send rovers to the cold traps to see if there really are thick sheets of relatively pure ice.

However a rover prospecting in cold traps would likely need nuclear power since solar panels aren't an option on these crater floors. Also a relay communications sat would be needed since the cold traps don't enjoy line of sight with the earth.

Cold trap prospecting is well beyond your team's abilities.

So you're right that Team Phoenix can't help with prospecting efforts. But you're wrong that any and all prospecting is a fool's errand.