The recent catastrophic loss of a satellite on SpaceX’s Falcon 9 rocket during ground testing reminds us that launching things into space is inherently dangerous. The other side of the difficulty coin is that spaceflight is also expensive. After all, accelerating tens of tons of complex machinery along an extremely precise path at a precise velocity requires all the technological knowledge and skill that we possess—and even then, we sometime fail to reach the exact destination desired. An article of faith among the space community is that lowering the cost of launch is an essential pre-requisite to the “opening” of the space frontier. Thus, SpaceX and many other companies are pursing techniques and technologies designed to recover and re-use rockets after launch. After all, we don’t throw away an ocean liner after it’s made a single voyage.
For many reasons, achieving the goal of low-cost launch service has been elusive. For one thing, the flight regime between Earth’s surface and low Earth orbit (LEO) is the most difficult one in which to operate. An atmosphere with highly variable conditions, high gravity, hypersonic speeds (both to and from space) and a possible watery dunking upon return are all potential roadblocks to the development of a successful reusable flight system. Some forget that the Space Shuttle (retired 5 years ago) was a largely reusable launch system—the Shuttle orbiter (with three, complex cryogenic engines used for launch) returned to Earth and the two solid rocket boosters were recovered (falling back to Earth over the ocean) and reused after each flight. However, over the course of a 30-year program, we found that refurbishing the vehicle and preparing it for re-flight was a much more costly labor- and time-intensive activity than had been anticipated. In the case of the Shuttle, reusability was not an asset, but an anchor—the continuing high costs to maintain and operate the Shuttle system left little funding for anything else the agency wanted to pursue, like human missions beyond LEO.
Still, the lure of reusable systems continues to fascinate—it just seems so logical. Yet, our path into space is constrained by what my friend Don Pettit calls the “Tyranny of the rocket equation.” This equation, first derived by Russian astronautics pioneer Konstantin Tsiolkovsky, simply means that for a rocket to deliver a payload to space, it must consist almost entirely of fuel (propellant), usually more than 90 percent of the mass of a rocket designed to send payloads to orbit. The “non-propellant” 10 percent fraction of that mass includes not only the payload (i.e., the object that you want to get into orbit) but also the airframe, shrouds, tanks, plumbing, avionics and other items critical to the proper functioning of the system. Combining highly combustible propellant and complex equipment is a formula for activity that is both “difficult” and “dangerous.”
But things get even more complicated. For human missions, we not only need to bring all of the equipment, computers and life support systems needed to keep us alive and to navigate through and operate in space, we need to bring large quantities of “dumb mass” —fuel, as well as air to breathe, water to drink, and food to eat. “Smart mass” has high information density (think complex machines, computers and people) while “dumb mass” has low information density (e.g., water, propellant, shielding). A rocket doesn’t know or care what kind of mass it launches—launching 100 kg of flight computers or 100 kg of water costs the same and we reach LEO with empty fuel tanks. To go beyond LEO into deep space, we need more propellant—and that’s a lot of dumb mass to drag up there with us from the bottom of the deepest gravity well in the inner Solar System (i.e., Earth’s surface).
There are three ways to deal with this issue. First, you can lower the cost of launching stuff into space—thus, the push for reusability. If you don’t throw away your vehicle after one use, it should lower your operating costs, which should permit launching more mass per dollar. However, this is not a completely straightforward solution because, as experience with the Space Shuttle system shows, the cost of reusability may end up being more than anticipated.
To illustrate, besides the cost of developing a reusable flight system for the Falcon 9 first stage (including landing legs, thrusters, flight software and a supporting ground system), making it reusable requires refurbishing the previously used rocket engines. The combustion of kerosene and oxygen creates carbon residue, requiring careful cleaning of each engine (Falcon has 9 engines in its first stage). It is not yet known if such refurbishment will require disassembly of each engine (increasing the number of labor hours and thus, costs). SpaceX says that by re-using the Falcon 9 first stage, the cost of a single flight will drop by “around 30%.” But this remains just an estimate, as no previously flown stage has yet been reused. In fact, no one will know exactly how much money will or can be saved by this innovation until several dozen flights of a “reusable” launch vehicle have taken place and we can understand just how much expense is incurred in preparing and re-using the boosters. For now, our history of reusability with the Space Shuttle advises caution on such claims.
A second approach to lowering cost is to launch large quantities of material at one time. This is one argument for the development of a “heavy lift” vehicle—by sending up more mass with each launch, we lower the cost per unit mass (i.e., dollars per kilogram) for space access. Both commercial and government programs are pursuing this idea for missions beyond LEO. SpaceX is developing its “Falcon Heavy” concept (a cluster of three Falcon 9 rockets, all firing simultaneously) to launch about 53 metric tons into orbit while NASA is developing its “Space Launch System” which can put about 70 metric tons into space. Both systems require multiple launches of these heavy lift vehicles to conduct human missions to the Moon (2-4) or to Mars (8-12), most of which would simply carry propellant for the journey beyond LEO.
But perhaps even this approach is barking up the wrong tree. What if we could get some (or a lot) of what we need not from Earth, but from space? Provisioning missions in space—beyond Earth’s deep gravity well—would lower the cost of spaceflight by establishing greater operational independence from Earth for humans beyond LEO. Which brings us to the third major way to lower costs—use space resources. Certainly the low-information “dumb mass” needed to support human life and for conducting missions into deep space can be obtained from space-based sources. We already get our energy for spacecraft locally in space—from solar arrays that collect sunlight and generate electricity. It’s time to accept and embrace the reality that we must obtain and use the abundant material resources of space. The Moon, asteroids and planets all offer material resources, the most valuable of which is water. Water can be converted through well-understood techniques into a variety of different and useful forms, including liquid oxygen and hydrogen—the most powerful chemical rocket propellant known.
To permanently lower the costs of human spaceflight, we should use a mixture of all of these techniques, but only where each one makes the most sense. For initial steps beyond LEO, a heavy lift launch system can deliver critical pieces (required early in a campaign) to the right localities with the minimum number of required launches. Reusability makes the most sense for space-based assets, like cislunar transfer stages, refueling depots, and reusable landers for the Moon and Mars. Because these systems are used only in the benign environment of space, they are subject to less destructive forces and corrosive interactions with Earth’s atmosphere and hydrosphere. By finding and harvesting water from the polar ice deposits of the Moon, we can power a permanent, sustainable, space-based transportation system. The use of lunar resources permits rapid build-up of space capability by relieving the launch system of the requirement to deliver tons of “dumb mass.” Like most difficult and seemingly intractable problems, lowering the cost of human spaceflight is not amenable to simple, quick or cheap solutions, but rather to a balanced, realistic, long-term, incremental approach that matches a technique’s strengths and advantages to the various required jobs that make up a complex, yet affordable, spaceflight system.