The Reverse Water Gas Shift

The reverse water gas shift (RWGS) reaction has been known to chemistry since the mid 1800's. While it has been discussed as a potential technique for Mars propellant manufacture in the literature4, there has been no experimental work done to demonstrate its viability for such application to-date. The RWGS reaction is given by equation (3).

CO₂ + H₂ = CO + H₂O           Delta H= +9 kcal/mole           (3)

This reaction is mildly endothermic and will occur rapidly in the presence of an iron-chrome catalyst at temperatures of 400 C or greater. Unfortunately at 400 C the equilibrium constant K_p driving it to the right is only about 0.1, and even at much higher temperatures K_p remains of order unity. There is thus a significant problem in driving the RWGS reaction to completion.

However, assuming that reaction (3) can be driven as written, an "infinite leverage oxygen machine" can be created by simply tying reaction (3) in tandem with the water electrolysis reaction (2). That is, the CO produced by reaction (3) is discarded while the water is electrolyzed to produce oxygen (the net product), and hydrogen which can be recycled to reduce more CO₂. Since all the hydrogen is recycled, barring leakage losses this can go on forever allowing the system to produce as much oxygen as desired. The only imported reagent needed is a small amount of water which is endlessly recycled.

The RWGS/electrolysis oxygen machine shares many of the advantages, and indeed can share many of the subsystem components, of an SE system. The RWGS reactor itself is just a simple steel pipe filled with catalyst, much like a Sabatier reactor, except that the catalyst is different. A similar condenser and identical water electrolysis system is also employed. Because the RWGS reaction is only mildly endothermic (9 kcal/mole for RWGS compared to 57 kcal/mole for water electrolysis), system power requirements are dominated by the water electrolysis step, the available technology for which is highly efficient. Moreover, since the thermal power required by the RWGS is less than that produced by the Sabatier reactor and their operating temperatures are comparable, a Sabatier reactor can be used to provide the heat required to drive the RWGS reactor. That is, if a Sabatier reactor running at a rate of 1 unit of equation (1) is lain side by side in direct thermal contact with a RWGS reactor running at a rate of 2 units of equation (3), the net reaction of the combined system will be:

3CO₂ + 6H₂ = CH₄ + 4H₂O + 2CO           Delta H= -22 kcal/mole           (4)

"Reaction" (4) is thus exothermic, requiring no net input power to operate. When run in combination with reaction (2), the net result is to produce 4 kg of methane and 16 kg of oxygen for every kg of H₂ imported, for a net propellant leverage of 20/1 and a O₂/CH₄ mixture ratio of 4/1. The energy efficiency of the combined RWGS/SE system is essentially the same as that in a simple SE system. Achieving such performance in a Mars in-situ propellant production system would be superb. The trick, however, is to find a practical way to drive the RWGS reaction to completion. There are a number of ways that this could be accomplished. These are:

a) Overload the reactor with CO₂ to force the complete consumption of H₂, and then recycle the excess CO₂ in the exhaust stream back into the reactor.

b) Overload the reactor with H₂ to force the complete consumption of the CO₂, and then recycle the excess H₂ in the exhaust stream back into the reactor.

c) Operate a system that removes water vapor from the reactor, thereby driving reaction (3) to the right. Such a system could either be a desiccant bed or condensing apparatus.

d) Combine approaches (a) and (c).

e) Combine approaches (b) and (c).

[CO][H₂O]/[CO₂][H₂] = [CO][0.01]/[CO₂][H₂] = 0.1           (5)

or, since the system is stoichiometric, and [CO₂] = [H₂] = X

CO = 10X²                     (6)

Fig.1. Schematic of RWGS system used as an "infinite leverage" O2 machine. In the first year of the MIRUR program, the subsystem drawn inside the dotted lines will be built and tested.

If X=1, CO=10 (i.e. the reactor is operating at 12 atmospheres, or 176 psi) and the conversion rate is 90%. This can be increased by going to higher pressures or increasing the ratio of H₂ to CO₂ in the input stream. The results for various reactor pressures and for both stoichiometric and 2:1 off-stoichiometric input ratios are shown in Fig. 2.

The yields shown in Fig. 2 will only be approached asymptotically as the rate of flow through the condenser loop approach a rate infinitely faster than the reactor net output flow (i.e. Flow2>>Flow1). Of course, the faster the rate of reactor gas recycle through the condenser loop, the more heat loss will occur, and more heating and loop circulation pump power will be needed. Preliminary analysis indicates, that with efficient counterflow heat exchanger design, that Flow2/Flow1 ratios greater than 10 and possibly as high as 50 may be practical.

However, provided the waste hydrogen is recycled, it really doesn't matter too much whether the real yield in the reactor is 80% or 99%, because CO₂ is available in unlimited quantities on Mars. On the other hand, if the desire for engineering simplicity makes it necessary to eliminate the hydrogen recirculation loop from the system, then real reactor yields are very important. While calculations can be of great assistance in predicting what such yields would be, the system, combining considerations of both chemical equilibrium and kinetics, system geometry, reactor temperature profiles, catalyst activity and surface area, is so complex that an accurate performance assessment can only be done by experiment. No one has done such an experiment to-date. We propose to do it.

Fig. 2. Conversion efficiencies in a RWGS system where a 10 C condenser is used to peg water vapor pressure in the reactor at 0.01 atmospheres.

Use of RWGS Reactor to Produce Ethylene

The discussion so far has shown how a RWGS reactor can be used either as the sole component in a loop with an electrolyser as an "infinite-leverage oxygen machine" on Mars, or how it can be used in tandem with an SE based Mars in-situ propellant production system to increase the leverage of such a system from 10.3/1 to 20/1. In addition, it should be obvious that, operating without an electrolyser, a RWGS reactor can be used to leverage imported hydrogen into water on Mars (to augment crew consumables) with a mass leverage ratio of 9/1. However the RWGS reactor opens up additional remarkable possibilities.

Let's say we operate the RWGS reactor with an excess of hydrogen, but we do not recycle the waste hydrogen effluent. As a simplified example, assume that the H₂/CO₂ input ratio is 3/1, and that the CO₂ conversion rate is close to 100%. Then we have 3 units of H₂ and 1 unit of CO₂ going into the reactor, 1 unit of H₂O collected in the condenser, and 1 unit of CO and 2 units of H₂ leaving the reactor. The water is electrolyzed to produce product oxygen for the propellant tanks and hydrogen for recycle into the RWGS. The CO and H₂ mixture can then be fed as input into an ethylene reactor, where in the presence of a iron Fischer Tropsch catalyst they can be reacted in accordance with:

2CO + 4H₂ = C₂H₄ + 2H₂O           Delta H=-49.4 kcal/mole           (7)

A schematic showing how reaction (7) could be operated in series with a RWGS reactor is shown in Fig. 3.

Fig.3. Schematic of RWGS/ethylene system. In the second year of the MIRUR program, the subsystem drawn inside the dotted lines will be built and tested.

Reaction (7) is strongly exothermic, and so like the Sabatier reaction, can be used as a heat source to provide the energy needed to drive the endothermic RWGS. It also has a high equilibrium constant, making the achievement of high ethylene (C₂H₄) yields possible. However, this system has extraordinary advantages over a Sabatier reactor. In the first place, ethylene has only two hydrogen atoms per carbon, while methane has four. Thus using ethylene for fuel instead of methane cuts the hydrogen importation requirement in half. Again, the propellant leverage of a RWGS/ethylene system is nearly double that of a RWGS/SE system, which itself is nearly double that of a simple SE system. In fact, with propellant leverage so high, it may be possible to acquire the required amounts of hydrogen from Mars atmospheric water vapor without too great a power impact, eliminating the hydrogen importation problem altogether. In the second place, ethylene has a boiling point (at one atmosphere pressure) of -104 C, much higher than methane's boiling point of -183 C. In fact, under a few atmospheres pressure ethylene is storable at Mars average ambient temperatures, whereas methane's critical temperature is below typical Mars nighttime temperatures. Thus ethylene can be liquefied on Mars without the use of a cryogenic refrigerator, whereas methane cannot be. This cuts the required refrigeration power for a RWGS based ethylene/oxygen system about in half relative to that of an SE based methane/oxygen production system. It also greatly reduces the need to insulate the ethylene fuel tanks. In the third place, the density of liquid ethylene is 50% greater than liquid methane, allowing for the use of smaller and therefore lighter fuel tanks on Mars ascent vehicles or ground rovers employing ethylene instead of methane fuel. Fourth, an ethylene/oxygen rocket engine should have a specific impulse about two seconds higher than a methane/oxygen rocket¹⁹, thereby slightly increasing overall mission performance. Fifth, ethylene has many other uses besides rocket or rover or welding fuel. It is used as an anesthetic. It is also used as a ripening agent for fruits and as a means of reducing the dormant time of seeds. These features could prove very useful in a developing Mars base which is aiming for self-sufficiency.

However, beyond this, ethylene is extraordinarily useful as the basic feedstock for a range of processes to manufacture polyethylene and numerous other plastics. These plastics can be formed into films or fabrics to create large inflatable structures and well as to manufacture clothing, bags, insulation, and tires, among others. They can also be formed into high-density stiff forms to produce bottles and other watertight vessels, tableware and innumerable other small but necessary objects, boxes, and rigid structures of every size and description, including those that are both transparent and opaque. Lubricants, sealants, adhesives, tapes, can all be manufactured; in fact the list is nearly endless. On Earth, ethylene has been characterized as the basis of the plastics industry²⁰ that has revolutionized modern life since the 1950's. The development of an ethylene-based plastics manufacturing capability on Mars would offer similar enormous benefits in opening up all sorts of possibilities and capabilities necessary for the human exploration and settlement of the Red Planet.

It may be noted that if Reaction (7) is not narrowly catalyzed, it will also have side reactions yielding methanol (CH₃OH) and propylene (C₃H₆). The later is not a problem, as propylene would be a superior product to ethylene, both as a fuel and a plastic feedstock (to produce polypropylene). Small methanol yields are acceptable, as methanol is miscible with ethylene and propylene and a mixture of the three would still make good storable rocket fuel. Excessive methanol yields would be a problem, however, because methanol molecules have four hydrogens for every carbon atom. Thus if reaction (7) were to produce only methanol (as the extreme case), the net propellant leverage would be no better than an SE system, while the specific impulse offered by its product when utilized in a rocket engine would be lower. This need not occur, however, as experiments have shown that if properly catalyzed, the methanol yield of reaction (7) can be kept as low as 2% by weight ²⁰.

Finally, it is clear that a RWGS based ethylene production system developed for space may have important application on Earth. Currently, the primary feedstocks for production of ethylene in the terrestrial chemical industry are ethane and ethanol, both of which have important other applications. In addition, natural ethane is a finite fossil fuel resource. In the RWGS/ethylene process, the basic feedstocks for ethylene production are water and CO₂. Thus, employing this cycle, relatively storable fuel (ethylene is liquid at 10 C at 735 psi) can be produced, which, when burned adds a net contribution of zero to atmospheric CO₂ and thus makes no contribution to global warming. If RWGS/ethylene production systems can be built which are even close to being economically competitive, the strong environmental benefits offered may make such systems quite attractive.

Summary of Applications and Advantages of RWGS Systems over the State of the Art

To summarize, the development of the RWGS system has many advantages over the state of the art for numerous applications that support NASA's Strategic Objectives in the areas of the Human Exploration and Development of Space, Scientific Research, and Space Technology. These applications include:

1. The ability to manufacture any amount of oxygen on Mars to support human exploration and robotic sample return missions. The only competing system that can do this is zirconia electrolysis. RWGS should be able to do it with a much more rugged and reliable system, on a much larger scale (if desired), with a power consumption about an order of magnitude less. If CO should be desired as a fuel, RWGS has to potential to produce it at least an order or magnitude more efficiently than a zirconia-electrolysis system.

2. RWGS reactors can also be used in tandem with electrolysis units to provide physical- chemical life support for oxygen regeneration and CO₂ disposal on space stations, Lunar bases, or piloted spacecraft anywhere in space. Compared to zirconia-electrolysis such a system is much more rugged and efficient. Compared to an SE based life support system, it has the advantage of wasting no hydrogen, and thus no water. Compared to a Bosch reactor based life support system, no solid graphite wastes are created.

3. RWGS reactors offer the ability to leverage imported hydrogen into water on Mars with a mass leverage of 9/1. Using a Sabatier reactor for this purpose would only produce a leverage of 4.5/1. Using a Bosch reactor would give 9/1 leverage, but would also produce solid graphite wastes that would be difficult to manage.

4. Used as an adjunct to a SE Mars in-situ propellant system, the RWGS reactor increases net propellant leverage from 10.3:1 to 20:1. This reduces tankage size and mass, and makes the hydrogen importation requirement for the system tractable.

5. Used as the front end of an RWGS/ethylene reactor system, the RWGS enables construction of a Mars in-situ propellant production unit which produces a high-energy propellant combination with a net leverage as high as 31/1. This is more than triple the leverage of a state of the art SE system. Moreover, the fuel produced is both denser than methane and storable on Mars without refrigeration.

6. The product ethylene can be used on Mars for other applications than rocket, rover and welding fuel. It can also be used as an anesthetic, as an aid to crop production, and as the basic feedstock for the manufacture of plastics for structures, fabrics, implements, and many other uses.

7. The RWGS/ethylene system may have important terrestrial applications as a way to produce relatively storable fuel whose combustion adds nothing to overall atmospheric CO₂ concentrations.