METHANE AND OXYGEN PRODUCTION
By: Andrés Felipe Vega
The Mars Direct architecture utilizes the in-situ production of CH4/O2 bipropellant for Earth return and surface
The Mars Direct architecture utilizes the in-situ production of CH4/O2 bipropellant for Earth return and surface
mobility from the very first mission. It can do this, because unlike the case of the Moon, the processes required to produce propellant on the surface of Mars are simple and very well understood. In fact, all of the chemical processes used in the Mars Direct plan have been in large scale use on Earth for over a century.
First there is the acquisition of the required native raw material. Since the hydrogen component of the bipropellant mixture represents only about 5% of the total propellant weight it can be imported from Earth. Heavy insulation of tanks with multi-layer insulation (MLI) can reduce in-space boiloff of liquid hydrogen to less than 1% per month during the 6 to 8 month interplanetary transit without any requirement for active refrigeration. Since the hydrogen raw material is not going to be directly fed into an engine, it can be gelled with a small amount of methane to prevent leaks. Gelling of the hydrogen cargo will also reduce boiloff further (as much as 40%) due to suppression of convection within the tank8.
The only raw materials thus required from Mars are carbon
and oxygen. The atmosphere of Mars, as measured by the 2 Viking landers, is composed of 95.3% CO2, 2.7%
nitrogen, 1.6% argon, and trace quantities of water, oxygen, and carbon-monoxide9. Carbon and oxygen are thus the two most plentiful elements in the Martian atmosphere and can be acquired "free as air" anywhere on the planet. The atmospheric pressure measured at the 2 Viking sites varied over a Martian year between 7 and 10 mbar, with a year round average of about 8 mbar (6 torr) observed at the higher altitude Viking 1 landing site on Chryse Planitia. Pumps which can acquire gas at this pressure and compress it to a workable pressure of 1 bar or more were first demonstrated by the English physicist Francis Hawksbee in 1709. Even better pumps are available today10.
In order to insure quality control in the propellant production process, it is desired that no substances of unknown composition, to wit, Martian dust, be allowed to enter the chemical reactors. This can be accomplished by first placing a dust filter on the pump intake to remove the vast majority of the dust, and then compressing the CO2 to about 7 bar pressure. When CO2 gas is brought to this pressure and then allowed to equilibriate to ambient Martian temperature conditions, it will condense into the liquid state. Any dust which managed to evade the pump filters will then go into solution, while nitrogen and argon will remain gaseous and thus can be removed. If CO2 is then vaporized off the holding tank it will be distilled 100% pure, as all dust will be left behind in solution. Distillation purification processes working on this principle have been widely used on Earth since the mid-1700s, when Benjamin Franklin introduced a desalination device for use by the British Navy.
Once pure CO2 is obtained, the entire process becomes
completely controllable and predictable, as no unknown
variables can be introduced by Mars. Thus with the design of adequate quality control on the CO2 acquisition
process, the entire rest of the chemical production process can be duplicated on Earth under precisely the same conditions that will be present on Mars, and reliability guaranteed by an intensive program of ground testing. Very few of the other key elements of a manned Mars mission (engines, aerobrakes, parachutes, life support systems, on-orbit assembly techniques etc.) can in fact be made subject to an equivalent degree of advance testing. This means that, far from being one of the weak links in the chain of a Mars mission, the in-situ propellant process can be made one of the strongest.
Once the CO2 is acquired it can be rapidly reacted with the
hydrogen brought from Earth in the methanation reaction, which is also called the Sabatier reaction after the chemist of that name who studied it extensively during the latter part of the 19th century.
The Sabatier reaction is: CO2+4H2=CH4+2H20 (1)
This reaction is exothermic and will occur spontaneously in the presence of a nickel catalyst (among others). The equilibrium constant is extremely strong in driving the reaction to the right, and production yields of greater than 99% utilization with just one pass through a reactor are routinely achieved. In addition to having been in widescale industrial use for about 100 years, the Sabatier reaction has been researched by NASA, the USAF, and their contractors for possible use in Space Station and Manned Orbiting Laboratory life support systems. The Hamilton Standard company, for example, has developed a Sabatier unit for use on Space Station Freedom, and has subjected it to about 4200 hours of qualification testing. It is interesting to note that the Hamilton Standard SSF Sabatier units, which use a proprietary Ruthenium catalyst with a demonstrated shelf life of greater than 12 years, are sized to react about 3 kg of CO2per day, which is the full capacity required to perform the Mini-Mars Direct sample return mission.
The fact that the Sabatier reaction is exothermic means that no energy is required to drive it, and this in turn
implies that the limiting rate at which it can be made to proceed on Mars is the rate at which the CO2 feedstock
can be acquired. CO2 can be compressed and liquefied
out of the Martian atmosphere at an energy cost of about 0.08 kWe-hr/kg. With 100 kWe available to drive the
pumps, the manned Mars Direct mission can acquire all of the 33 tonnes of CO2 needed to completely react its initial
supply of 6 tonnes of liquid hydrogen into methane and water in just 26 hours. Similarly, with 0.8 kWe, the Mini- Mars Direct mission can acquire the 550 kg it needs to react away its hydrogen supply in about 55 hours. This would not actually be done in either mission, as it would lead to a needless oversizing of the Sabatier reactors and the pumps. The point, however, is that the hydrogen can be reacted away at a rate much higher than it will boiloff, and thus there is no problem with the long term storage of the cryogenic liquid hydrogen on the Martian surface.
As the reaction (1) is run, the methane so produced is liquefied either by thermal contact with the hydrogen stream or (later on after the liquid hydrogen is exhausted) the use of a mechanical refrigerator. (Methane is slightly less cryogenic than liquid oxygen.) The water produced is condensed and then transferred to a holding tank, after which it is pumped into an electrolysis cell and subjected to the familiar electrolysis reaction:
2H2O = 2H2 + O2 (2)
The oxygen so produced is refrigerated and stored, while the hydrogen can be recycled back to the Sabatier reaction (1).
Electrolysis is familiar to many people from high school chemistry, where it is a favorite demonstration experiment. However, this universal experience with the electrolysis reaction has created a somewhat misleading mental image of an electrolysis cell as something composed of Pyrex beakers and glassware strung out across a desk top. In reality modern electrolysis units are extremely compact and robust objects, composed of sandwiched layers of electrolyte impregnated plastic separated by metal meshes, with the assembly compressed at each end by substantial metal end caps bolted down to metal rods running the length of the stack. Such solid polymer electrolyte (SPE) electrolyzers have been brought to an extremely advanced state of development for use in nuclear submarines, with over 7 million cell-hours of experience to date. Testing has included subjecting cells to depth charging and loads of up to 200 g's. Both the Hamilton Standard and the Life Sciences companies have also developed light weight
electrolysis units for use on the Space Station. Once again, these units are of adequate capacity to perform the propellant production operation for the Mini-Mars Direct mission. The SPE units that Hamilton Standard has supplied for use by Britain's Royal Navy have the correct output level to support the propellant production requirements of the manned Mars Direct mission. These units have operated for periods of up to 28,000 hours without maintenance, about 4 times the utilization required on the manned Mars Direct mission. The submarine SPE electrolysis units are very heavy, as they are designed to be so for ballasting purposes. SPE electrolysis units designed for space missions would be much lighter (see below).
If all the hydrogen is expended cycling the propellant production process through reactions (1) and (2), then each kilogram of hydrogen brought to Mars will have been transformed into 12 kg of methane/oxygen bipropellant on the martian surface, with an oxygen to methane mixture ratio of 2:1. Burning the bipropellant at such a ratio would provide a specific impulse of about 340 s, assuming an nozzle expansion ratio of 100. This amount of propellant mass leveraging would be satisfactory for the Mini-Mars Direct mission, providing the sample returned was kept to 10 kg, and the rover was reduced from 400 to 350 kg, with an extra 50 kg of hydrogen taken compared to the estimate given in Table 1. However the optimum oxygen to methane combustion mixture ratio is about 3.5:1, as this provides for a specific impulse of 373 s and the hydrogen to bipropellant mass leveraging of 18:1. It is this level of performance is the basis of the optimal design of the manned Mars Direct mission, as well as the mass estimates for the MMD mission given in Table 1.
If this optimal level of performance is to be obtained, an additional source of oxygen must be obtained beyond that made available by the combination of reactions (1) and (2). One possible answer is the direct reduction of CO2.
2CO2 = 2CO + O2 (3)
This reaction can be accomplished by heating CO2 to
about 1100 C, which will cause the gas to partially dissociate, after which the free oxygen so produced can be electrochemically pumped across a zirconia ceramic membrane by applying a voltage. The use of this reaction to produce oxygen on Mars was first proposed by Dr. Robert Ash at JPL in the 1970s, and since then has been the subject of ongoing research by both Ash (now at Old Dominion University), Kumar Ramohalli and K. R. Sridhar (at the Univ. of Arizona), and Jerry Suitor (at JPL). The advantage of this process is that it is completely decoupled from any other chemical process, and an nfinite amount of oxygen can be so produced without any additional feedstock. The disadvantages are that the zirconia tubes are brittle, and have small rates of output so that very large numbers would be required for the manned Mars Direct application. (The numbers would not be excessive for the MMD mission.) Improved yields have recently been reported at the Univ. of Arizona, so the process may be regarded as promising, but still experimental.
An alternative that would keep the set of processes employed firmly within the world of 19th century industrial chemistry, would be to run the well known water-gas shift reaction in reverse. That is recycle some of the hydrogen produced in the electrolysis unit into a third chamber where it is reacted with CO2 in the presence of an iron- chrome catalyst as follows:
CO2+H2=CO+H2O (4)
This reaction is mildly endothermic but will occur at 400 K, which is well within the temperature range of the Sabatier reaction. It has been shown by Meyer11 that if reaction (4) is cycled with reactions (1) and (2), the desired mixture ratio of methane and oxygen can be produced with all the energy required to drive reaction (4) provided by thermal heat output from the Sabatier reactor. Reaction (4) can be carried out in a simple steel pipe, making the construction of such a reactor quite robust. The disadvantage of reaction (4) is that in the temperature range of interest it has an equilibrium constant of only about 0.1, which means that in order to drive it to the right it is necessary to both overload the left hand side of the equation with extra CO2 while condensing out water to remove it from the right hand side. This is certainly feasible, and actually constitutes a fairly modest chemical engineering design problem. However a number of alternatives that are at least equally promising have been advanced. One of the most elegant of these would be to simply combine reactions (1) and (4) in a single reactor as follows:
3CO2 + 6H2 = CH4 + 2CO + 4H2O (5)
This reaction is mildly exothermic, and if cycled together with reaction (2) would produce oxygen and methane in a mixture ratio of 4:1, which would give the optimum propellant mass leveraging of 18:1 with a large extra quantity of oxygen also produced that could function as a massive backup to the life support system. In addition, salvageable CO would also be produced that could conceivably used in various combustion devices or fuel cells. If all the CO and O2produced is included, the total propellant mass leveraging obtained could thus be as high
as 34:1. Researchers at Hamilton Standard intend to put to the test a number of methods of driving this reaction early in 199112.
Probably the easiest method of obtaining the required extra oxygen is just to take some of the methane produced in reaction (1) and pyrolyze it into carbon and hydrogen.
CH4=C+2H2 (6)
The hydrogen so produced would then be cycled back to attack more Martian CO2 via reaction (1). After a while a
graphite deposit would build up in the chamber in which reaction (6) was being carried on. (This reaction is actually the most common method used in industry to produce pyrolytic graphite.) At such a time, the methane input to the reactor would be shut off and instead the chamber would be flushed with hot CO2 gas. The hot CO2 would then react with the graphite to form CO, which would then be vented, cleaning out the chamber.
CO2+C=2CO (7)
Such a plan, incorporating two chambers, with one carrying out pyrolysis while the other is being cleaned, has been suggested to the authors as the simplest solution to the extra oxygen problem by a group of researchers at Hamilton Standard13.
The Hamilton Standard group also provided mass estimates for propellant production systems for both the Mars Direct and MMD missions based on a system combining reactions (1), (2), (6), and (7). The estimates are given in T able 2.
Table 2. Hamilton Standard Mass Estimates for CH4/O2 Plant
The mass estimates in Table 2. assume 2 complete units each with 100% mission capacity for the system used in the MMD mission, and 3 units each with 50% capacity for the system employed on the manned Mars Direct mission.
The reason for the different approach to redundancy on the two missions is that the small scale units used on the MMD mission have essentially the same mass whether they are full or half capacity, while the much larger units used on the manned Mars Direct scale in a roughly linear manner with capacity.
In summary the methods required to produce methane and oxygen bipropellants on Mars for the Mars Direct are well understood and already in an advanced state of technology development. It has suggested in some quarters that while these propellant production processes are promising, they should be relegated to inclusion in downstream missions, with the initial set conducted using only terrestrial propellants. This hardly seems appropriate, as these propellant production methods are in a more mature state of development than nearly everything else associated with the manned Mars mission, and except for the Titan IV/Centaur launch vehicle, everything in the MRSR mission as well. Moreover, the carrying out of an initial set of manned Mars missions without the leverage afforded by in-situ propellant production would be the ultimate in architecture incoherence, as the initial set of non-in-situ propellant missions would require a different set of vehicle hardware, and a massive launch and orbital infrastructure that would be very costly and later prove unnecessary. Furthermore, if much in the way of useful surface exploration is to be accomplished, the in-situ process will be needed anyway - and on the very first mission. Since we can have the propellant production process right from the start of the manned Mars exploration program, and since we must have it if useful surface exploration is to be done, we might as well take full advantage of it and use it to provide the Mars ascent and Earth return propellant as well.
